This distinction offers a useful lens for thinking about when to use AI. If the process is tedious but the outcome matters, AI can take over the drudgery while a human oversees the result. If the output has no audience and no impact, AI simply makes it cheaper to produce meaningless work. Understanding the difference is the core challenge for anyone adopting AI in a professional setting.

LinkedIn Discovers Slop

LinkedIn recently introduced a button that lets users flag posts as “seems like AI slop.” The platform also removed its “enhance your post” feature, which had been using AI to help compose posts, and replaced it with a proofreader that preserves the author’s voice. This is a welcome change, but it raises a question: why are people producing these posts in the first place? Many feel obligated to post for “thought leadership” and career advancement, even when they have nothing original to say.

AI can generate 500 words on how coaching a daughter’s soccer team taught seven lessons about enterprise procurement. The post will be polished and may even attract likes, but it will sound exactly like everything else in the feed. Readers want a personal point of view, something that makes the author human. Outsourcing that voice to a machine guarantees the opposite.

This does not mean all AI-assisted writing is bad. AI can help a non-native English speaker express an original idea, turn dictated thoughts into a coherent draft, challenge an argument, find missing evidence, or suggest a better structure. Many writers use AI for these purposes and produce excellent work. But if the writer has nothing to say, AI will not supply a real voice.

Just Because You Can

A few years ago, a friend who ran product marketing at a large technology company decided to generate all sales collateral with AI. First-call decks, email templates, and other materials were produced by machines. His logic was brutally rational: the collateral produced by hand was pulling only a few dozen downloads from a sales force of thousands. If almost nobody wanted it, why pay a junior marketer to create it? Let the machine publish into the void.

The problem was that he had not solved the foundational issue. Low download numbers could mean many things: sellers could not find the material, it was not useful at scale, or a few dozen people used it to close enormous deals. Maybe the wrong collateral was being created for a pressing need that sellers resolved on their own. Or maybe the thing that every product marketing team does simply does not need to be done.

AI makes it easier not to wrestle with these questions. A deck costs almost nothing to produce, so the company keeps making decks instead of asking whether they should exist. A weekly report takes minutes to generate, so it keeps being published. The knowledge base fills with pages no one reads because stopping a process requires a decision, while automating it only requires a prompt. AI becomes a way to kick the can down the road, avoiding the hard human decisions that professionals are paid to make.

The same pattern appears in software development. App creation is up dramatically, but app adoption is not. Building was never the only constraint. Getting anyone to care is the constraint, and AI does not solve that. It just removes the last excuse for not noticing.

Two Kinds of Low-Value Work

This is not an argument against automating boring tasks. Automating boring, repetitive work is a great idea. But “low-value work” hides two very different things that deserve opposite treatment.

The first is a low-value process attached to a valuable outcome. Expense reports, backups, and other administrative chores are not exciting, but they must get done. AI can handle as much of the process as safely possible, leaving a human to check the result and move on. This is the sweet spot for AI adoption.

The second is a low-value process attached to no discernible outcome. Sales collateral with no audience, LinkedIn posts with no personality, and knowledge base articles no one reads fall into this category. Automating this feels like a win because the cost drops, but cost was never the core problem. The real problem is that the output does not need to exist. If you cannot name a useful outcome that would be lost if the output stopped existing, you do not have an automation opportunity. You have a cancellation opportunity.

I Use AI Where I Care Most

The inverse is also true. The work where AI is used most aggressively is often the work people care about most. This complicates the argument, but it reveals an important principle.

One example comes from genealogy research. An AI model was asked to read four 18th-century probate wills as part of a 20-year search for a missing ancestor. The model fabricated an entire emigrant ancestor, complete with a clean and plausible story. The only way to catch the error was to click through to high-resolution images and read the documents line by line. The model excused its behavior as “hopeful reading,” but the real reason the error was caught was because the researcher cared deeply. Twenty years of caring made them open the originals and verify every detail.

The same pattern holds in the workplace. An executive memo is high-stakes work with a real reader, a real decision, and the author’s name on it. Using AI to compress a sprawling pile of data and argument into a concise recommendation is valuable, but only if the author reads every line, challenges the output, and edits heavily. AI does 90% of the early legwork, freeing the human to focus on the critically important last 10%.

This is the AI trust tax: you pay the verification cost up front, or someone else pays it later when the answer causes damage. The tax is paid most reliably when somebody cares about the outcome. When nobody cares, AI can make bad work look finished enough to ship and presentable enough to glance at without reviewing carefully.

AI slop, therefore, is not a model problem. It is a question of caring, which shows up in how people choose to use AI.

Maybe Just Stop

There is no single guiding principle that works for every situation, but two questions can help decide when to use AI and when to walk away.

• What useful outcome disappears if this work stops?
• If that outcome matters, where is human judgment still needed?

If there is no meaningful outcome, stop doing the work, whether AI-powered or human-powered. If the outcome matters but much of the process does not, hand the process to AI. Let AI buy the groceries, search the archive, assemble the first draft, or build the sales deck that sellers have actually requested. Then keep a person accountable for the result.

AI is extremely good at making more things. But the world does not need more things. Making them cheaper will not make them useful. Humans, by contrast, are extremely good at determining which things need to be made. That is the job of every professional, and no prompt can take it away.

Source: InfoWorld News

The numbers are hard to ignore. US technology companies, including Alphabet, Amazon, Meta, and Microsoft, are expected to spend about $650 billion on AI-related infrastructure in 2026, up from roughly $410 billion in 2025, according to industry analysis. That kind of growth tells us something important. AI is not just another software wave that sits neatly atop the existing cloud stack. It is forcing a redesign of the stack itself.

That redesign reaches deep into networking and data movement. Nvidia recently announced plans to invest $2 billion each in photonics companies Lumentum and Coherent, underscoring where the pressure points are emerging. The issue is no longer only raw compute. It is also how quickly data can move between processors, racks, and clusters without creating unacceptable bottlenecks or power inefficiencies. As AI systems scale, latency, throughput, and energy usage become first-order economic concerns.

All of this suggests that AI will drive more demand for public cloud computing, but it will do so unevenly. Public cloud providers remain the fastest way to access advanced infrastructure, global scale, and managed AI services. At the same time, the cost profile of large, persistent AI workloads is prompting many enterprises to reconsider whether the traditional hyperscaler model should remain the default destination for every stage of the AI life cycle.

Most AI starts in the public cloud

When companies are experimenting, speed matters more than optimization. Public clouds give teams immediate access to GPUs, foundation model APIs, vector databases, orchestration tools, security controls, and integration services. They also allow businesses to quickly start pilots without waiting for procurement cycles, data center expansions, or specialized infrastructure teams.

Given the high level of uncertainty, the public cloud is often the right choice for first-generation AI. Enterprises do not yet know which use cases will deliver value, how much inference traffic they will see, or which architecture model will ultimately survive. At this stage, the ability to quickly try many things is more important than squeezing every dollar from the underlying infrastructure. Managed services reduce friction, and friction is the enemy of early adoption.

This is why we are seeing strong initial demand for AI to land in public cloud environments. Enterprises are building chatbots, copilots, knowledge assistants, document automation systems, and code generation tools there because the cloud dramatically lowers the barrier to entry. It provides compute as well as a full operating environment for AI experimentation.

Next-gen AI systems present choices

The second generation of enterprise AI systems looks different. Once a use case proves its value and usage becomes persistent, the financial model changes. A workload that looked inexpensive during a proof of concept can become shockingly expensive when it runs at production scale, especially if it depends on premium GPU instances, high-performance storage, constant network traffic, and managed services layered on top of one another.

That is where repatriation enters the conversation. We are starting to see a pattern in which enterprises build first-generation AI systems on public clouds, learn what works, and then move some of those workloads back on-premises or onto so-called neocloud providers that offer AI-optimized infrastructure at a lower cost.

On-premises deployment is attractive when utilization is steady, data gravity is high, governance requirements are strict, and the organization has sufficient scale to justify owning or directly controlling the infrastructure. Neocloud options become attractive when enterprises still want an external provider but do not want to pay the full premium often associated with large hyperscalers. These specialized providers are increasingly positioning themselves around dense GPU capacity, simpler pricing, and architecture built specifically for AI rather than for general-purpose enterprise IT.

This is an important adoption pattern because it dispels the old assumption that cloud migration is always one-way. In the AI era, workload placement is becoming more fluid. Enterprises are learning that the best place for experimentation may not be the best place for steady-state production and that AI economics can punish architectural laziness much faster than traditional enterprise applications ever did.

AI and public cloud demand

How much demand will AI drive for public cloud computing? Quite a lot, especially in the near term. Every major enterprise AI initiative will likely engage the public cloud in a meaningful way, whether for model development, training bursts, integration services, security tools, or global deployment. But it would be a mistake to assume that all demand will remain locked in traditional hyperscalers over time.

Some AI workloads will stay in the public cloud permanently because they are bursty, globally distributed, hard to predict, or tightly coupled to cloud-native services. Other workloads, especially those with stable usage patterns and heavy inference volume, will be candidates for relocation. Economics will drive those decisions more than ideology.

The likely outcome is a more segmented market. Public clouds will dominate the front end of AI adoption and continue to play a major role in hybrid operations. On-premises environments will regain relevance for cost-sensitive, steady-state, and compliance-heavy workloads. Neocloud providers will grow as a middle option for enterprises seeking external AI capacity without paying full hyperscaler prices. In short, AI will increase public cloud demand, but it will also heighten scrutiny of the correct fit in the long term.

Three factors to consider

First: Speed and cost are distinct metrics. The public cloud is usually the fastest way to get an AI initiative off the ground, and that speed has real business value. But the architecture that wins a pilot may end up destroying the production budget. Enterprises need a placement strategy from day one, even if they start in the cloud.

Second: AI workload economics differ from those of traditional applications. Training, inference, data movement, storage, and model serving can interact in ways that quickly create cost surprises. Organizations should model not only compute usage but also utilization patterns, network flows, and the costs of managed services surrounding the core AI stack. Without that discipline, they risk designing systems that are technically elegant but financially unsustainable.

Third: Future flexibility matters more than short-term convenience. Enterprises should avoid building AI systems so tightly around a single provider's proprietary stack that moving becomes painful or impossible. The winners in this market will be the companies that preserve optionality, enabling them to shift workloads across public clouds, on-premises environments, and emerging neocloud platforms as economics, regulations, and business requirements evolve.

The real question is not whether the cloud will benefit, but how long each AI workload will remain in the cloud. AI will unquestionably generate significant new demand for public cloud computing. For most enterprises, AI workloads will stay in the cloud long enough to enable rapid innovation, but they will not necessarily remain there forever.

Source: InfoWorld News

Economic forces at work

The timing is not accidental. Homes are expensive, especially for those who bought at the elevated prices and interest rates of late. Mortgage payments are a heavy burden; insurance and taxes continue to climb. In this housing market, homeowners are increasingly interested in turning underutilized parts of their properties into sources of recurring income. Spare rooms have become short-term rentals. Garages have become workshops or accessory units. Rooftops have become solar assets. Now, major players in the housing market are considering basements, utility rooms, and detached structures as potential spaces for small-scale server infrastructure.

At the same time, businesses are under pressure to rethink where compute lives. AI is increasing the demand for processing capacity. Edge workloads continue to grow. Not every application needs to run in a hyperscale facility, and not every business wants to pay for hyperscale economics. There is a strategic appeal to pushing workloads closer to users or into lower-cost, more widely distributed locations. Residential hosting becomes one possible answer to a question the industry is already asking: How much infrastructure can be decentralized without losing economic and operational control?

There is also a cultural shift at work. More technically capable homeowners now understand racks, uninterruptible power supply systems, network monitoring, remote access, and even local power upgrades. The old gap between enterprise infrastructure knowledge and prosumer infrastructure knowledge has narrowed. That makes the idea feel more achievable, even if the barriers to doing it commercially remain substantial.

Business models taking shape

The most important point to understand is that there is not yet a large, polished market in which random homeowners openly host random third-party servers the way people list rooms on Airbnb. What does exist are several adjacent business models that point in that direction without fully embracing the concept of residential colocation.

One model is the controlled edge-host program. In this arrangement, a company places or manages compute equipment in selected distributed locations, often with strict standards for connectivity, power, and maintenance. The homeowner or site operator is not acting as an open colocation provider. Instead, they participate in a curated hosting network where the provider controls the service architecture.

Another model is the decentralized compute marketplace. These platforms allow individuals or smaller operators to sell spare compute capacity from their own hardware. This is closer to the economics of monetizing residential infrastructure. Still, it is not the same as taking custody of someone else’s physical server and being responsible for the environment in which it runs. Selling compute cycles is one thing. Housing enterprise hardware is another.

A third model is the traditional infrastructure broker or marketplace. These companies already match buyers and sellers for colocation, bare-metal, and related services. They are proof that brokering infrastructure relationships is a viable business. But those relationships generally connect enterprises to professional facilities, not to homeowners willing to make room for a small server farm next to their furnace or water heater.

In other words, the components of a market are visible. Distributed demand exists. Brokering exists. Willing hosts likely exist. But the residential version remains incomplete because the trust, standardization, and liability models are still underdeveloped.

The upside is obvious

The strongest positive component of this potential market is its financial aspect. If a homeowner can generate enough monthly income to offset part of a mortgage payment, the idea will always attract attention, especially in newer housing markets, where monthly carrying costs are high, and people are seeking durable sources of supplemental income. Hosting infrastructure sounds like, at least in theory, a more stable and less socially intrusive way to monetize a property than opening a home to a constant stream of short-term tenants.

There is also an argument for asset utilization. Many homes contain underused spaces that could produce some economic return. A basement corner, a detached workshop, or a dedicated utility room may be worthless from a revenue perspective until someone turns it into something productive. If infrastructure providers are willing to pay for access to space, power, and connectivity, the home begins to function as part of the digital economy rather than simply as shelter.

For businesses, the appeal is equally straightforward. Residential locations may offer lower real estate costs, faster deployment, and better geographic distribution for select workloads. In regions with relatively inexpensive electricity and strong connectivity, a modest amount of residential hosting could fill gaps that do not warrant full commercial data center expansion. Homes will not replace data centers; rather, they might, in a very narrow set of circumstances, complement them.

The downsides are everything else

The problem with the whole idea is that the negatives are significant. Residential power is not data center power. Residential broadband is not enterprise-grade networking. A private home is not a secure, redundant, environmentally controlled facility, no matter how carefully a rack is installed.

Power is the first issue. Most homes are not designed to handle sustained commercial server loads without electrical upgrades. These upgrades can be expensive, heavily regulated, and dependent on local utility cooperation. Once backup batteries, uninterruptible power supply systems, cooling equipment, and dedicated circuits are added, the project starts to look less like a side hustle and more like a facilities operation.

Heat and noise follow quickly. Commercial hardware generates both continuously, which affect the comfort of the house, the cost of climate control, and the long-term reliability of the equipment. It also transforms residential life. Maintenance becomes routine. Monitoring becomes constant. The house begins to absorb the rhythm of an always-on machine room.

Then come the risks that stall many otherwise creative ideas. Fire hazards. Water damage. Physical theft. Tampering. Insurance complications. Zoning restrictions. HOA objections. Lease restrictions for tenants. Questions about who can access the equipment and when. Liability if a customer’s hardware is damaged. Compliance concerns if sensitive data or regulated workloads are involved. All of these factors are manageable in theory, but they are precisely why professional facilities exist.

Customer trust may be the biggest obstacle of all. Most businesses are comfortable buying compute from a recognized provider because they assume a predictable operating environment. That assumption weakens significantly when the infrastructure sits in a private residence. Who is responsible during an outage? What happens if there is a storm, a flood, or a neighborhood power event? How is physical access controlled? How are incidents documented? Those questions are not edge cases. They determine the model’s viability.

What is realistic from here?

Residential data hosting is unlikely to become the next mainstream large-scale hosting model. The economics of professional data centers still win in most situations because those facilities were built to solve exactly the problems that home models will struggle to address. Reliability, security, redundancy, and customer assurance are difficult and expensive to achieve. Purpose-built environments handle them better.

Still, the concept should not be dismissed outright. In some parts of the country, there may be a path forward. Cheap power. Upgradeable electrical service. Strong broadband. Detached or isolated space. Favorable local rules. Workloads that benefit from geographic distribution and do not require pristine enterprise conditions. In those scenarios, carefully managed micro-hosting could make sense.

That is probably the realistic future. Not an Airbnb for random servers. Not whole neighborhoods that are converted into basement data centers. Instead, a selective market where curated providers match specific homeowners or small properties with specific infrastructure needs under tightly controlled terms. What will start as a niche could still be enough to matter.

Source: InfoWorld News

Superlogical, a startup founded by HashiCorp co-founder Mitchell Hashimoto, aims to solve that problem by turning the terminal into a persistent execution layer that both developers and AI agents can share. The company is proposing a server-side multiplexer that keeps software sessions alive across devices and environments, allowing humans and AI agents to resume, share, and manage long-running coding sessions without losing context. Rather than simply improving the terminal emulator itself, Superlogical rethinks where session state lives and how it is synchronized.

A terminal that remembers

Traditional multiplexers like tmux were created to preserve terminal sessions for individual users. They allow developers to detach from a session, reconnect after a network interruption, and continue from the same screen. That functionality was revolutionary when remote development over unreliable connections was the norm. But tmux and its contemporaries were built for a single human operator staring at one terminal window. They have no concept of an AI agent that might be waiting for approval, executing a long-running build, or pausing to inspect test results.

That limitation is becoming a real pain point for developers and enterprises adopting long-running AI coding agents, according to analysts. “Tools like tmux were built for a human watching one terminal, but agents now run for hours in the background, and existing tools have zero awareness of whether an agent is waiting for your approval or still thinking,” said Manoj Chandra Jha, principal analyst at Nord-IQ Research. Developers often have to manually reconnect to sessions simply to determine whether work is progressing, blocked, or ready for review. This becomes increasingly cumbersome as enterprise teams begin running multiple autonomous coding agents simultaneously.

Why AI agents need more than a terminal

Modern AI coding agents do not simply execute a single command and exit. They can analyze codebases, edit files, run tests, debug failures, and iterate on solutions over extended periods. An agent may need to fetch context from a database, interact with a remote API, or wait for a human to approve a risky refactoring. During that time, the session must remain alive and coherent, preserving environment variables, working directories, shell history, and application state.

Terminal emulators like Ghostty, also created by Hashimoto, have improved rendering performance and user experience, but they still operate on a client-server model that assumes the server is the source of truth and the client is a passive viewer. Superlogical flips that model by moving the authoritative session state to the server. The server stores the full session context, including processes, environment, and terminal output, while clients just render the raw data stream.

Instead of having both the server and client repeatedly parse and render terminal output, the new multiplexer stores the authoritative session state on the server and streams raw terminal data to clients, which independently reconstruct the same session using Hashimoto’s libghostty rendering engine. This means a desktop client, a web browser, or a mobile app can all connect to the same session and see exactly the same screen, with no drift or desynchronization.

Resuming and sharing sessions

For developers, the practical benefit is clear. You can start a complex debugging session on your workstation, close your laptop, and later resume the same session from a tablet or a cloud-based IDE. The session is not a snapshot or a log; it is a live execution context that persists through network changes, device switches, and even user logouts.

For AI agents, the benefit is even greater. Agents can now be launched from one interface and monitored from another. A developer could start an agent from a mobile phone on the train, let it run in the cloud, and later attach to the session from a desktop terminal to review the agent’s work. The agent itself does not need to know where the user is or which device they are using. It simply writes to the shared session, and the user can connect from anywhere.

This approach, Jha said, will allow developers and AI agents to reconnect to, share, and resume long-running sessions from desktop, mobile, or web clients while preserving the same execution context. That means reduced interruptions, simplified remote development, and improved productivity for cloud-native and AI-assisted software engineering, echoed Pareekh Jain, principal analyst at Pareekh Consulting.

Bringing control to autonomous software work

For CIOs and enterprise architects, however, the appeal is not just developer productivity. It is also about control and observability. Right now, AI agents often run in isolated environments with no consistent audit trail. A developer might invoke an agent from a CLI, the agent might make changes to a remote repository, and those changes might be reviewed or deployed without any centralized record. This fragmentation makes it difficult for organizations to answer basic questions: What exactly did the agent do? Who authorized it? Which files were changed and why?

Superlogical’s persistent session model could help address those concerns. By making a standardized session the unit of work, the platform can track exactly what happened, who did what, and how sessions are handed off between human and AI. “Right now, AI agents run in the background with basically no audit trail. A standardized session that tracks exactly what happened, who did what, and hands off cleanly between human and AI could make it much easier to monitor and secure all this new agent activity,” Jain said.

The approach could also help enterprises gain a single control layer for tracking, auditing, and managing autonomous software work instead of relying on multiple monitoring, observability, and workflow tools. Jha added that Superlogical could become the connective tissue between the terminal, the agent orchestrator, and the enterprise security stack.

Early-stage technology with enterprise ambitions

Still, analysts warn that Superlogical is not yet a mature infrastructure platform. The product remains in beta and has yet to be independently benchmarked or proven at enterprise scale. There are open questions about security, session isolation, multi-tenancy, and how well the server-side multiplexer performs under heavy load with many concurrent sessions and agents.

Another concern is compatibility. Enterprises have invested heavily in existing terminal workflows, CI/CD systems, and remote development tools. Superlogical must integrate with that ecosystem rather than replace it. The company will need to support standard terminal protocols, shell environments, and popular editors without requiring a complete infrastructure overhaul.

Hashimoto’s track record with HashiCorp gives Superlogical credibility among infrastructure engineers. HashiCorp built a suite of tools, including Vagrant, Packer, Consul, Terraform, Nomad, and Vault, that became foundational to modern DevOps practice. Hashimoto later shifted his focus to Ghostty, a fast, native terminal emulator that quickly gained attention for its performance and clean implementation. Superlogical is the next logical step in his exploring how to bridge the gap between human-centric development tools and AI-driven software automation.

The underlying idea is that terminal sessions are not just a display mechanism; they are a record of work. For decades, that record was ephemeral and personal. If a developer closed a laptop, the session was either gone or locked to a specific machine. tmux made sessions detachable, but only for a single user and only with the assumption that the user would return to the same terminal or an equivalent one.

Superlogical argues that the next generation of software development needs sessions that are first-class objects, shared across people, machines, and AI agents. This is not just about keeping a process running after a network timeout. It is about making the execution context itself a persistent, auditable, and shareable resource.

Whether Superlogical ultimately becomes that persistent execution layer or remains a more capable alternative to tmux will depend on whether it can prove those benefits at enterprise scale, the analyst added. If it succeeds, it could reshape how enterprises deploy, observe, and govern AI coding agents. If it stalls, it may still be a useful tool for individual developers, but with a much smaller impact than its founder envisions.

Source: InfoWorld News

On Thursday, Dili announced a $15 million Series A round, bringing the company’s total capital raised to $21.7 million. The round follows a $6.7 million seed round and positions Dili to expand its platform beyond its current base of customers. The company’s software is already being used on about 700 projects, according to co-founder and CEO Anand Chaturvedi, and the new funding will help the company scale its operations and refine its compliance engine.

An unusual bet on infrastructure compliance

The Series A was led by Khosla Ventures, a firm known for backing ambitious artificial-intelligence companies. Participating investors included Allianz, Rebel Fund, Brick and Mortar Ventures’ Darren Bechtel, and Y Combinator’s Garry Tan. Dili was part of Y Combinator’s Summer 2023 batch, giving the startup access to a broad network of founders and operators in the construction and energy sectors.

“AI for compliance” has become a common pitch among startups, but Dili focuses on a narrow and particularly messy slice of the market: the rules that govern construction projects, especially those receiving some form of federal funding. That focus is deliberate. The infrastructure boom tied to AI data centers, the Inflation Reduction Act, and reshoring of manufacturing has created an urgent need for tools that can handle wage rules, safety regulations, and environmental requirements at scale.

Why compliance is so difficult

To understand the problem, consider the rules that apply to a single federally funded project. Under the Davis-Bacon Act, the Department of Labor sets prevailing wages for workers on certain public works jobs. A separate set of prevailing wage and apprenticeship rules, often called PWA rules, applies to clean-energy projects funded under the Inflation Reduction Act. Additional OSHA workplace-safety requirements and EPA environmental rules may apply depending on the nature of the work. Each of these regimes has its own definitions, reporting schedules, and penalties.

• Davis-Bacon requires contractors to pay wages set by the Department of Labor for specific trades and regions. The requirement extends to subcontractors and often covers projects that receive federal funding.
• PWA rules add apprenticeship quotas and certification standards, typically requiring a certain percentage of labor hours to be performed by registered apprentices.
• OSHA and EPA rules vary by project type and can impose separate record-keeping and inspection obligations.

The overlapping requirements make manual compliance difficult. A compliance team might normally sample a subset of payroll records, invoices, and apprenticeship documents, hoping to catch problems before they become costly. Dili, by contrast, checks all the information as it comes in. “Non-compliance can result in millions of dollars of fines for those projects,” Chaturvedi explained. “So it’s really powerful to be able to check all the information as it comes in, instead of just sampling data.”

Keeping AI in its lane

Given those high stakes, reliability is a central concern. Chaturvedi believes Dili’s architecture prevents any LLM-based fuzziness from sneaking into the final product. Contemporary AI models are only used in the company’s data layer, the engine that takes unstructured documents and translates them into structured data. From there, a deterministic system sorts the data according to the complex-but-static compliance rules. The distinction is important for customers who need an auditable, repeatable answer for regulators.

“Imagine being able to read across the entire context of a company’s internal documents, all of their vendors’ documents, all of their ERP information, all of their payroll systems information, and then draw out the data that you need specifically for reporting or compliance,” Chaturvedi said. The AI layer is designed to make that possible without forcing every subcontractor onto a single software platform. If a vendor sends a PDF, a spreadsheet, or even a scan of a physical document, the data layer can extract the relevant fields.

The deterministic rule engine then applies the legal requirements. For example, if a worker’s classification is pulled from a payroll record, the engine can check it against the prevailing wage schedule for that trade and location. If the wage is too low, the system flags the discrepancy immediately. If an apprentice certification is missing, the project manager sees the issue before a regulator does. This approach turns a compliance task that might take a full day of manual review into a process that can be completed in minutes.

Infrastructure boom and the compliance bottleneck

The need for such a system has grown alongside the infrastructure boom. AI data centers require massive amounts of power, which has led to a surge in new energy projects, transmission lines, and battery-storage facilities. Many of those projects are built with the help of federal tax incentives, meaning the contractors must meet the labor standards attached to those incentives. At the same time, reshoring and supply-chain diversification have pushed manufacturers to break ground on new plants, many of which also rely on federal support.

Construction has historically been slow to adopt digital tools. Payroll records arrive in different formats from different subcontractors, and a general contractor often has to reconcile paper documents with ERP systems and union reports. Dili’s technology is built for that patchwork. By making the data layer AI-driven, the company can handle variation in document types and formats without forcing every subcontractor onto a single compliance suite.

Deployment and business model

Dili is already making the system work in practice. Chaturvedi said the software is being used at about 700 projects, ranging from manufacturing facilities to data centers. Notably, roughly half of those projects use Dili as an in-house software tool, while the other half outsource the entire compliance process to the company on a contractor model. Dili is able to handle both types of contract, although Chaturvedi anticipates the industry will shift more toward the software model in the years to come.

“Software and AI are going to start eating a lot of those professional services workflows, so I think more and more people will start to bring those in-house,” Chaturvedi said. “The interesting thing will be how the market itself evolves and where the customer needs go as AI develops.”

The company’s growth suggests that the market is ready for that shift. Between the seed round and the new Series A, Dili has raised $21.7 million, with strong support from investors who have deep ties to both AI and industrial construction. The company’s Y Combinator pedigree and its focus on a concrete, high-value problem give it a clear path as federal infrastructure spending continues to expand.

For project owners, contractors, and subcontractors, the takeaway is that compliance is becoming an automated process. The tools are still early, but the trajectory is clear: AI will be used to extract facts from messy documents, and deterministic software will apply the rules. That combination is already reducing the time and cost of compliance on hundreds of projects, and it is likely to become the standard for the next generation of infrastructure development.

Source: TechCrunch News

The deal is expected to close in the third quarter of Okta&8217;s fiscal 2027, subject to customary closing conditions. That timeline suggests the company is moving quickly to integrate Permiso&8217;s technology and team ahead of what industry analysts expect to be a surge in enterprise adoption of AI-driven workflows and machine identities.

Why Okta is buying Permiso

Okta&8217;s move to buy Permiso comes as identity management companies seek to expand beyond verifying users at login to continuously monitoring what users, applications, and AI agents do once they gain authorized access to a network environment. That shift has intensified competition to secure machine identities as enterprises embed AI deeper into everyday operations.

Traditional identity and access management tools have largely focused on authentication at the perimeter&8212;ensuring that the person or service requesting access is who and what they claim to be. But with the rise of cloud-native architectures and AI agents that operate autonomously, security teams increasingly need visibility into what happens after access is granted. Permiso addresses this gap by providing identity threat detection and response, or ITDR, capabilities that continuously analyze activity across cloud environments and detect anomalous behavior that could indicate a compromised identity or a malicious insider.

By adding Permiso&8217;s capabilities, Okta is positioning itself to offer a more comprehensive identity security fabric that spans the entire lifecycle of an identity, from initial authentication to ongoing behavior monitoring. This is particularly important for AI agents, which are often granted broad permissions to execute tasks and may be exploited by attackers to move laterally within an organization&8217;s infrastructure.

Permiso&8217;s origins and technology

Permiso, which emerged from stealth in 2022, develops software that helps security teams spot suspicious activity in cloud environments after users or applications have been granted access. More recently, the startup has expanded its platform to monitor AI agents and other machine identities.

The company was co-founded by Paul Nguyen and Jason Martin, both former FireEye executives. FireEye is known for its advanced threat intelligence and incident response capabilities, and Nguyen and Martin brought that expertise to bear on the identity security problem. Permiso specializes in detecting attacks that use stolen or compromised identities to move through cloud infrastructure, a technique often referred to as “identity-based attacks.”

In April, the startup introduced SandyClaw, a platform designed to analyze AI agent skills in a sandboxed environment to identify malicious behavior before those agents are deployed into production. This is a growing concern for enterprises: AI agents are being given access to sensitive data and systems, and if those agents are compromised or maliciously configured, they could cause significant damage. SandyClaw aims to give security teams a way to test AI agents without putting their organization at risk.

Permiso&8217;s technology is built on a foundation of deep visibility into cloud identity behavior. It can ingest data from a wide range of sources, including cloud service provider logs, identity provider logs, and application logs, then apply machine learning and analytics to identify anomalies. This approach enables security teams to respond to threats in real time rather than after the fact, which is critical in a landscape where attackers are increasingly using automated tools to compromise identities.

Funding and valuation

Permiso has raised about $29 million to date, including an $18.5 million Series A round in April 2024 led by Altimeter Capital. People familiar with the financing said the Series A valued the Palo Alto-based startup at about $80 million on a post-money basis. That means the $200 million acquisition price represents a significant return for investors, reflecting the strategic premium that Okta is willing to pay to gain a foothold in the emerging AI identity security market.

The company&8217;s previous funding includes a seed round that was not publicly announced in detail, but the combination of seed funding and the Series A has been sufficient to build out a robust product portfolio and attract enterprise customers. Permiso&8217;s customer base is said to include large organizations in the technology, financial services, and healthcare sectors, though the company has not publicly disclosed a comprehensive customer list.

Okta&8217;s product integration plans

Okta&8217;s chief product officer, Ely Kahn, said in a prepared statement that the acquisition will advance the company&8217;s threat detection and prevention capabilities. “Permiso will extend Okta&8217;s identity security fabric with proven identity threat detection and response capabilities, and an incredible threat research and security team that will advance Okta&8217;s threat detection and prevention capabilities,” Kahn said.

We expect that Permiso&8217;s technology will be integrated into Okta&8217;s existing identity cloud platform, which already offers a range of solutions including single sign-on, multi-factor authentication, and lifecycle management. The addition of Permiso&8217;s ITDR capabilities will likely enable Okta customers to monitor for suspicious activity across their AWS, Azure, and Google Cloud environments, as well as on-premises infrastructure, all from a single console.

Okta has been investing heavily in AI-driven security features in recent quarters. The company has rolled out machine learning-based risk scoring and adaptive authentication capabilities, but Permiso brings a more specialized and enterprise-grade set of detection and response tools. By folding Permiso&8217;s threat research team into Okta, the company will gain the ability to stay ahead of emerging attack techniques that target machine identities and AI agents.

The broader market for AI identity security

The identity security market is undergoing a significant transformation as organizations increasingly rely on non-human identities, such as service accounts, APIs, and AI agents. Some industry analysts estimate that machine identities now outnumber human identities by significant margins in large cloud environments, and each of those machine identities represents a potential vector for attack.

Traditional security tools were not designed to handle the scale or behavior patterns of machine identities. For example, an AI agent that is tasked with processing invoices may interact with dozens of other services and databases, creating a complex web of activity that would be difficult for a human security analyst to review manually. Automating the monitoring and analysis of these interactions is essential, and that is exactly what Permiso&8217;s platform is built to do.

Okta is not the only company moving into this space. Major cloud providers and security vendors have been expanding their own identity threat detection offerings, and startups in the ITDR space have attracted significant venture capital investment in recent years. But Okta&8217;s acquisition of Permiso gives it a differentiated position because it combines the breadth of its identity management platform with the depth of Permiso&8217;s specialized detection engine.

Furthermore, the timing of the deal appears strategic. Many enterprises are currently in the early stages of deploying generative AI and AI agents across their operations, and they are realizing that securing these new workloads is different from securing traditional applications. AI agents often require broad permissions to perform tasks, and they may act in ways that are difficult to predict. This creates a need for security controls that can adapt to changing behavior patterns and that can detect when an agent has been hijacked or misused.

Permiso&8217;s SandyClaw product is particularly well-suited to this challenge, as it allows organizations to simulate AI agent behaviors in a controlled environment and analyze the underlying skills that agents are equipped with. This can help identify malicious or flawed capabilities before they cause harm, effectively serving as a security testing tool for AI deployments.

With the acquisition, Okta is also gaining a team of threat researchers who have deep experience in identity-based attacks and cloud security. This team will likely work alongside Okta&8217;s existing security research organization to develop new detection rules and threat intelligence feeds for customers.

As the deal awaits regulatory and shareholder approval, Okta is positioning itself to be a leader in what is becoming a critical area of enterprise security. The company has traditionally been known for its customer-facing identity products, but this acquisition signals an ambition to be a deeper security provider, one that protects the entire identity infrastructure of an organization, including the rapidly growing number of machine identities and AI agents.

Source: TechCrunch News

The Apple Deals roundup is designed as a one-stop destination for anyone looking to save on Apple hardware and accessories. Whether you are shopping for a Mac, iPad, iPhone, Apple Watch, or add-ons such as cases, cables, docks, bags, and software, this guide brings together the latest discounts and price comparison tools in one place.

Deals are added and removed regularly, so the page is updated throughout the week as new offers are discovered. Shoppers are encouraged to bookmark the page and check back often to catch limited-time discounts. The weekly section also highlights significant markdowns on Macs, iPhones, iPads, and Apple Watch models, along with giveaways and app sales.

Below the deals section, price comparison charts cover all of Apple's Macs and iPads. The charts compare prices from five major Apple resellers: Best Buy, Amazon, MacMall, Adorama, and B&H Photo. Prices highlighted in green indicate especially strong discounts, making it easy to spot the best possible deal at a glance. Readers who discover an exceptional deal can send it in for potential inclusion in future roundups.

Weekly Deals

Each week, the roundup highlights the best discounts on Apple-related accessories, including cases, bags, docks, Apple Watch bands, hard drives, and more. Recent deals have included back-to-school bargains on MacBook Pro, AirPods, Apple Watch, and AirTag 2; Best Buy's Back to School Sale; and Amazon discounts on AirPods, AirTag 2, and iPads. Other notable offers have included $100 off AirPods Max 2, Apple Watch Series 11 starting at $299, MacBook Pro savings up to $500, and a sitewide Nomad anniversary sale with up to 30% off.

Deals are updated throughout the week to ensure that the discounts listed remain current. The goal is to provide a constantly changing selection of bargains, from everyday accessories to high-end Apple hardware, so that both casual shoppers and Apple enthusiasts can find meaningful savings.

Deals Guides

For shoppers who want a deeper dive into a specific product line, dedicated Best Deals guides cover the current best prices for AirPods, HomePod, Apple TV 4K, MacBook Pro, MacBook Air, MacBook Neo, iMac, iPad, and iPhone. These guides are designed to make the shopping process faster and easier by gathering all relevant price information in a single place.

MacBook Neo

The MacBook Neo is Apple's new budget laptop, equipped with the A18 Pro chip from the iPhone 16 Pro. It is available in four colors and starts at $599 for 256GB of storage, with a 512GB configuration and Touch ID power button available for a $100 upgrade. The notebook features a 13-inch display, a 1080p FaceTime HD camera, two USB-C ports, a 3.5mm headphone jack, and two side-firing speakers.

Apple rates the MacBook Neo for up to 16 hours of battery life, and the machine weighs just 2.7 pounds. It also includes a Magic Keyboard and a Multi-Touch trackpad. In the price comparison charts, the best prices for the 256GB model have recently ranged from about $689 to the regular $699, while 512GB models have been seen from around $785.

MacBook Air

The MacBook Air remains Apple's most affordable notebook, combining cost, performance, and portability in a thin design. In March 2026, Apple updated the 13-inch and 15-inch MacBook Air with its latest M5 chip, featuring a 10-core CPU and what Apple calls the world's fastest CPU cores. The chip also includes up to 10 GPU cores with Neural Accelerators, delivering up to four times faster performance for AI tasks than the previous M4-based model.

The M5 MacBook Air adds Apple's custom N1 wireless chip for Wi-Fi 7 and Bluetooth 6 connectivity. Storage now starts at 512GB and is configurable up to 4TB. In recent price comparisons, some 13-inch M5 models have dropped to around $1,199, while 15-inch models have been seen at $1,399, making the Air a strong option for value-focused buyers.

MacBook Pro

The MacBook Pro is Apple's most powerful notebook, and the latest update brings M5 Pro and M5 Max chips. The 14-inch MacBook Pro with M5 Pro starts at $2,199, while the 16-inch version starts at $2,699. The 14-inch M5 Max model starts at $3,599, and the 16-inch M5 Max model starts at $3,899. Both new chips use a Fusion Architecture that bonds two third-generation 3nm dies into a single chip with advanced packaging.

The M5 Pro and M5 Max chips feature an 18-core CPU, up from the 14-core and 16-core designs of the previous generation. Apple says the 16-inch model can last up to 24 hours on a single charge and can fast-charge to 50 percent in about 30 minutes with a 96W or higher USB-C adapter. Price comparisons show frequent discounts on both 14-inch and 16-inch models, with some 14-inch M5 Pro configurations available for around $2,259.

iMac

Apple's iMac is an ultra-thin all-in-one desktop that was refreshed with the M4 chip in October 2024. The M4 iMac comes in a range of colors and includes a new nano-texture display option, along with significant speed and efficiency improvements. Apple also raised the base memory to 16GB, with configurations available up to 32GB of RAM and up to 2TB of SSD storage.

The current 24-inch iMac models start at $1,299 for the 8-core M4 chip with 256GB of storage, while higher-end 10-core M4 configurations with 512GB or 1TB storage are available at higher price points. Recent deals have brought some configurations down to around $1,424 or $1,459, depending on the retailer and color choice.

Mac mini

The Mac mini is Apple's most compact desktop, designed for users who already have a display, keyboard, and mouse. Apple introduced a redesigned M4 and M4 Pro Mac mini in October 2024, marking the first major design change for the product in over a decade. The M4 models start at $599, making the Mac mini one of the most affordable entry points into the Mac ecosystem.

Retailers have been offering the M4 Mac mini with 16GB of RAM and a 256GB SSD at various prices, with some configurations listed near $799. The M4 Pro models start higher and are aimed at professionals who need additional performance for demanding workflows.

Mac Pro

The Mac Pro is Apple's desktop machine for professionals, and the current model was introduced in June 2023 with the M2 Ultra chip. It features a 24-core CPU, support for up to a 76-core GPU, and up to 192GB of memory. The tower design includes eight Thunderbolt 4 ports, two high-bandwidth HDMI ports, seven PCIe expansion slots, dual 10 Gigabit Ethernet ports, and a headphone jack.

The Mac Pro starts at $6,999, and custom configurations can push the price significantly higher. Because of the wide range of build options, shoppers are encouraged to compare pricing across retailers for specific configurations before making a purchase.

Mac Studio

Apple announced the Mac Studio in March 2022 as a powerful companion to the Studio Display, and the third-generation version arrived in March 2025 with M4 Max and M3 Ultra chip options. The chassis remains similar to a larger Mac mini, with two front ports and an SD card slot, while the rear offers four Thunderbolt 5 ports, HDMI, 10-Gigabit Ethernet, two USB-A ports, a headphone jack, and a power button.

The Mac Studio starts at $1,999. When paired with the 27-inch Studio Display, it serves as a modern replacement for the 27-inch iMac as Apple's high-end desktop solution. The Studio Display features 5K resolution, up to 600 nits of brightness, P3 wide color support, and an A13 Bionic chip for advanced camera and audio features.

Apple Watch

The Apple Watch is available in three main models: Series 11, SE 3, and Ultra 3. The Series 11 is the flagship device, offering hypertension detection, Sleep Score, 5G connectivity, and a full set of health sensors including heart rate and ECG. The SE 3 is a lower-cost alternative with an S10 chip, always-on display, Double Tap support, wrist temperature sensing, and sleep apnea detection.

The Apple Watch Ultra 3 targets athletes and explorers with satellite connectivity, 5G cellular, and the largest display Apple has ever put in a watch. Recent deals have brought GPS Series 11 models down to around $299, while SE 3 models have been available from about $219. Ultra 3 models have also seen discounts of up to $100 or more on select bands.

Apple TV

The Apple TV 4K is Apple's set-top box for bringing the Apple ecosystem to the television. The current model, updated in October 2022, includes an A15 Bionic chip, HDR10+ support, and the new Siri Remote. The 64GB base model is priced at $129, while a 128GB version adds an ethernet port and support for Thread networking.

While the hardware has not been updated in a few years, the Apple TV 4K remains a popular choice for those invested in Apple services. Retailers often offer small discounts on both storage tiers, with the 128GB model frequently seen at around $249.

iPad Pro

The iPad Pro lineup includes 11-inch and 13-inch models, both updated in October 2025 with the M5 chip. The tablets feature Ultra Retina XDR OLED displays, an optional nano-texture display, and an extremely thin design. The 11-inch model is 5.3mm thick and weighs less than a pound, while the 13-inch model is Apple's thinnest product ever at 5.1mm.

The M5 chip brings improved AI performance, faster storage, and better memory bandwidth. Apple's custom N1 networking chip adds Wi-Fi 7 support, and cellular models use the C1X chip for up to 50 percent faster cellular data speeds. The 11-inch iPad Pro starts at $999, and the 13-inch model starts at $1,299. Recent price comparisons show select 256GB Wi-Fi models available for as low as $1,139 for the 11-inch and $1,422 for the 13-inch.

iPad Air

In March 2026, Apple introduced a new iPad Air with the M4 chip, available in 11-inch and 13-inch sizes. The tablet sits between the entry-level iPad and the iPad Pro, offering many Pro-like features at a lower price. It includes a 12-megapixel Center Stage front camera, 12-megapixel rear camera, Touch ID power button, USB-C with up to 10 Gbps data transfer, Wi-Fi 7, and Bluetooth 6.

The 11-inch iPad Air starts at $749, while the 13-inch model starts at $949. The device supports the Apple Pencil Pro and comes in four colors. Shoppers looking to save even more may find discounts on previous-generation 2024 iPad Air models, which are often available at lower prices once a new generation arrives.

iPad

Apple updated the entry-level iPad in March 2025 with an A16 chip, delivering up to 30 percent faster performance than the previous model. The base storage was doubled to 128GB, and new 256GB and 512GB configurations were added. The 128GB Wi-Fi model starts at $349, with four color options available.

The iPad is a popular choice for students and families who want Apple's tablet experience at a budget price.

Source: MacRumors News

Phone App: Call Context

The biggest update to the Phone app in iOS 27 is Call Context. This new feature uses Apple Intelligence to surface relevant information from the Mail app when you call a company. Instead of having to dig through emails to find a confirmation code or reservation number, the Phone app does the work for you behind the scenes.

For example, if you call an airline, the Phone app might show your confirmation code for an upcoming flight. If you're calling a hotel, it could display your reservation code or check-in details. The system doesn't listen to your conversation or analyze what you're saying. Instead, it looks at the phone number you're calling and matches it against details detected in the Mail app, then presents the most likely relevant information on the call screen.

Call Context is an Apple Intelligence feature that runs entirely on-device. That means all the analysis happens on your iPhone, and your data doesn't get sent to Apple servers. This privacy-first approach is consistent with Apple's broader strategy for on-device intelligence, which has become a major selling point for recent iPhones.

Because of the hardware requirements for on-device Apple Intelligence, Call Context requires an iPhone 15 Pro or newer. That includes all iPhone 16 and iPhone 17 models, as well as the iPhone Air. Interestingly, if you make a phone call from your iPad or Mac, Call Context works there too, as long as you're signed in with the same Apple ID and running compatible software. That cross-device support makes it useful for people who rely on their Mac or iPad for work and prefer to take calls there.

Call Context is a natural evolution of features Apple has been building for years. It's reminiscent of the smart suggestions in apps like Calendar and Reminders, but it's far more context-aware. Rather than just showing a phone number, it gives you the critical details you likely need before the person on the other end even answers. For frequent travelers or anyone who juggles multiple appointments, this could be a genuine time-saver.

FaceTime: Dual Capture

FaceTime also has a couple of new features in iOS 27, but the most exciting is Dual Capture. On the iPhone 17 models and the iPhone Air, you can now use the front and back cameras at the same time while on a FaceTime call. This allows the person you're talking to see both your face and whatever is behind you, whether you're showing off a new purchase, giving a tour of a room, or trying to get help with something you're looking at.

Dual Capture works by splitting the video feed so that both the front-facing and rear-facing cameras are active simultaneously. The person you're calling sees your face in one portion of the screen and the rear camera feed in another. It's a bit like the Picture-in-Picture mode that's popular in streaming apps, but for live video calls.

Turning on Dual Capture is simple. On a supported iPhone, you just tap the camera flip button during a FaceTime call. If you want to share only the back camera feed, you can tap to turn off the front camera view. This gives you flexibility to switch between sharing your face, sharing your surroundings, or both at the same time.

There are some hardware limitations to be aware of. Dual Capture video streaming is limited to the iPhone 17, iPhone 17 Pro, iPhone 17 Pro Max, iPhone Air, and newer models. But the receiving device doesn't need any special hardware. As long as the person you're talking to is running iOS 27, they can see dual video streams on a FaceTime call. So if you have an older iPhone, you can still receive and watch a dual video feed, even if you can't send one yourself.

This feature opens up a lot of possibilities. For example, a parent could read a bedtime story to a child while also showing the book's illustrations through the rear camera. Someone working on a DIY project could show a friend the instructions while also asking for advice, and the friend would be able to see both the project and the person's reactions simultaneously. It's a small change on the surface, but it makes FaceTime calls feel more like a shared experience.

Apple has been slowly expanding FaceTime's capabilities over the years, and Dual Capture is arguably the most significant visual addition since FaceTime gained support for video effects and screen sharing. It also puts FaceTime on par with some third-party video calling apps that have offered simultaneous camera feeds for a while, though Apple's integration is more seamless and doesn't require any extra configuration.

Other FaceTime Improvements

In addition to Dual Capture, iOS 27 brings several other FaceTime upgrades that, while less flashy, still contribute to a better calling experience.

Connection Quality

Apple says FaceTime calls will now have better quality when you have a bad connection. This is likely thanks to improved encoding and error detection algorithms that can adapt to fluctuating network conditions. In practice, it means the video and audio may stay smoother and clearer even when your signal is weak. This type of under-the-hood improvement is easy to overlook, but it's crucial for anyone who uses FaceTime to stay in touch with family or colleagues in areas with spotty cellular coverage.

Wi-Fi and Cellular Transitions

iOS 27 also makes iPhones better at transitioning from a Wi-Fi network to a cellular connection and vice versa. If you move out of range of your home Wi-Fi while on a FaceTime call, the iPhone will now switch to cellular data more smoothly, so you're less likely to experience a dropped call or a long pause. The same applies in reverse when you come back into range of a trusted Wi-Fi network. This feature is especially useful for people who walk around while talking, or who commute through areas with variable network coverage.

Live Captions in Traditional Chinese

Live Captions, which were already available in several languages, now include Traditional Chinese support in iOS 27. This is a meaningful addition for Chinese-speaking users who rely on captions to follow FaceTime calls or who prefer to read along while listening. Traditional Chinese is widely used in Taiwan, Hong Kong, and Macau, so expanding Live Captions to this language helps make FaceTime more accessible to a larger audience.

Communication Safety Expanded

Apple is also expanding Communication Safety, the feature that helps protect users, especially children, from receiving sensitive content. In iOS 27, Communication Safety now blocks gore and violence in FaceTime and Messages, in addition to the nudity protection it already offered. This means the system can detect and blur images or videos that contain explicit violence before they're viewed, and it can provide helpful resources and options to the person receiving them.

This expansion reflects Apple's continued focus on giving parents and users more control over the kinds of content they interact with. While nudity detection was already a significant step, broadening the scope to include gore and violence makes the feature much more comprehensive. It's a thoughtful addition for parents who want to give their kids a safer online experience without having to constantly monitor their devices.

iOS 27 Launch Details

iOS 27 is currently being beta tested. It's available to developers and public beta testers, who have been exploring its new features and helping Apple iron out bugs. The final version of iOS 27 will launch this fall, alongside the new iPhone models that are expected to be announced in September.

This year's update is designed to refine the overall experience rather than completely overhauling every app. Apple has leaned heavily into Apple Intelligence, with a revamped Siri that's more context-aware and capable of handling complex tasks. But it's also focused on stability and performance, fixing bugs and improving system responsiveness.

For iPhone users, iOS 27 brings dozens of small but welcome changes across the system, but the Phone and FaceTime updates are among the most practical for everyday use. Call Context and Dual Capture are the kind of features that you don't realize you need until you try them, and the improvements to connection quality and network transitions make the core calling experience better whether you're on a FaceTime call or a traditional cellular call.

As with any major iOS release, not every feature will be available on every device. Apple's habit of reserving certain capabilities for newer hardware means that users with older iPhones may miss out on some of the highlights. But the software is designed to be backwards-compatible where possible, so even if you're not upgrading to a new iPhone this year, you'll still see improvements in FaceTime and the Phone app.

For those who do own a supported iPhone, the combination of Call Context and Dual Capture makes iOS 27 feel like a meaningful upgrade. It's clear that Apple is thinking about how people actually use their phones to communicate, and these features are designed to reduce friction and make conversations more productive and personal.

In the coming weeks, Apple will likely continue to refine these features as the beta period progresses. By the time iOS 27 ships this fall, the Phone and FaceTime apps should be polished and ready for prime time. If you're curious about the new capabilities, the public beta is a good way to get a preview, though you should always back up your device before installing beta software.

With these updates, Apple is making sure that the most basic functions of an iPhone – making calls and video chatting – remain modern, intelligent, and secure. It's a reminder that sometimes the most important features aren't the ones that make headlines, but the ones that make your daily life a little bit easier.

Source: MacRumors News

September 2026

• iPhone 18 Pro: The iPhone 18 Pro is expected to feature a smaller Dynamic Island and a variable aperture main camera lens that works more like a DSLR. It will likely run on the A20 Pro chip built using Apple's 2nm process, and include Apple's C2 modem. Rumored colors include dark cherry and blue.
• iPhone Fold / iPhone Ultra: Apple's first foldable iPhone is rumored for fall 2026. It may have a 5.3-inch external display that unfolds into a 7.6-inch internal display, with a book-style design similar to an open iPad mini. The device could be super thin at 4.5mm when open, and is said to feature a highly durable hinge that nearly eliminates the crease. It may use an A20 chip, dual rear cameras, and a Touch ID sensor built into the power button instead of Face ID. Price estimates suggest it could cost as much as $2,500.
• Apple Watch Series 12: The next Apple Watch is not expected to feature major design changes. No new capabilities have been rumored yet, so it could be primarily a chip refresh.
• iPad mini 8: The iPad mini 8 is expected to adopt an OLED display, marking a major technology upgrade. It may also feature a more water-resistant design with a vibration-based speaker system that removes external speaker holes. The device could use the same A20 chip as the iPhone 18 Pro.
• HomePod mini 2: The updated HomePod mini is reportedly ready, but its launch depends on the new version of Siri. It will feature an updated S-series chip based on the Apple Watch Series 10, and new color options are expected. Some rumors mention Apple's N1 networking chip, though leaked code suggests otherwise.
• Home Hub: Apple's upcoming home hub is expected to support smart home controls, music playback, video calls, weather, Siri requests, photos, notes, and calendar events. It will feature an all-display design with a 7-inch screen. Two models are anticipated: one wall-mountable and one with a speaker base resembling a HomePod mini.
• Apple TV 4K: The next Apple TV 4K will likely include a new Apple-designed Wi-Fi and Bluetooth chip, plus a faster A17 Pro processor. This should enable console-quality gaming and Apple Intelligence. No major design changes are expected.
• Apple Smart Home Camera: Apple is reportedly designing its own indoor security cameras that connect via HomeKit and interface with the planned home hub. These could arrive as soon as late 2026.

Late 2026

• 14-inch MacBook Pro: The entry-level 14-inch MacBook Pro is expected to receive the M6 chip, making it one of the first machines with Apple's 2-nanometer processor technology.
• iMac: The 24-inch iMac will likely be updated with the M6 chip as well. No design or display changes are expected, so it will continue to resemble the 2024 model.

Late 2026 to Early 2027

• OLED MacBook Pro: A MacBook Pro with an OLED touchscreen display could launch in late 2026 or early 2027. The higher-end 14-inch and 16-inch models are expected to use M5 Pro and M5 Max chips, since M6 Pro and M6 Max chips are not planning to arrive in that timeframe. The OLED MacBook Pro is expected to have an updated design.
• Mac Studio: A refreshed Mac Studio with M5 Max and M5 Ultra chips is in development, but release timing remains unclear due to RAM shortages and high component costs.
• Mac mini: Both M5 Pro and M6 Mac mini models are in development. Launch timing is uncertain, with possible release windows in 2026, early 2027, or later in 2027.

2027 Launches

• iPhone 18: Apple is reportedly shifting to a split launch strategy. The standard iPhone 18 may not arrive in September 2026; instead, it could launch in spring 2027. The fall 2026 event would focus on the iPhone 18 Pro, iPhone 18 Pro Max, and the first foldable iPhone.
• AirPods Max: A refreshed AirPods Max is expected in 2027, with a lighter design. It could include the H2 chip or newer, bringing improved sound quality, better active noise cancellation, longer battery life, Adaptive Audio, Personalized Volume, Conversation Awareness, and support for "Siri" instead of "Hey Siri."
• iPad Pro: No new iPad Pro is expected in 2026. In 2027, however, it may receive a faster M6 chip and vapor chamber cooling for enhanced performance.
• AirPods Pro: The next AirPods Pro could include at least one tiny infrared camera and a next-generation H3 chip with improved sound quality and lower latency. These cameras might support in-air gestures, especially for Vision Pro integration. The product may still be called AirPods Pro 3.
• iPhone 18e: Slated for spring 2027 alongside the iPhone 18, the iPhone 18e will likely feature minor changes.
• iPhone Air 2: A second-generation iPhone Air is reportedly delayed from fall 2026 and could arrive in spring 2027. Apple is said to be considering adding a second camera to increase appeal, and the device might be more affordable than the original.
• iPad 12: An updated low-cost iPad is in development, reportedly featuring the newest A19 chip that debuted in the iPhone 17. No other major features or design changes are known at this time.
• 20th Anniversary iPhone: The 2027 iPhone will reportedly have an enclosure that curves around the device edges, eliminating bezels. It may use a narrow metal band around the middle instead of a full metal frame, and the selfie camera will move under the display. This would be the first iPhone with no display cutouts at all. It is expected to launch in September 2027.
• Tabletop Robot (J595): Apple's tabletop robot with a thin robotic arm and swivel base is expected to be a more powerful version of the home hub. It is rumored to launch sometime in 2027.
• MacBook Air: Refreshed 13-inch and 15-inch MacBook Air models are planned for early 2027, likely with the M6 chip.
• 14-inch MacBook Pro with M7: An updated entry-level 14-inch MacBook Pro with an M7 chip could arrive as early as 2027. It will not have an OLED display, but it is expected to share the same design overhaul as higher-end models.
• MacBook Neo: A new version of the low-cost MacBook Neo is coming, though timing is unknown. It is expected to feature a faster A19 Pro chip, more memory, and possibly new colors.

Products With Unknown Release Dates

• AirPods 5: Apple is developing AirPods 5 with an updated H3 chip for improved sound quality and lower latency. Two models are reportedly in the works, likely continuing the ANC and non-ANC split.
• Apple Watch Ultra 4: Development is underway, but no feature details or launch timing are known. Apple has not refreshed the Apple Watch Ultra on a strict annual schedule.
• AI Smart Glasses: Apple is working on smart glasses to compete with Meta's Ray-Bans. They are expected to include cameras, speakers, sensors, and AI, but no displays. The glasses would support photos, music, phone calls, and AI interactions. A 2026 introduction is possible, with a 2027 launch.

Late 2027 to Early 2028

• Second OLED MacBook Pro: A second-generation OLED MacBook Pro is expected roughly a year after the first release. These models will likely feature M7 Pro and M7 Max chips built for artificial intelligence workloads.

2028

• OLED MacBook Air: Apple plans to bring OLED displays to the MacBook Air, but 2028 is the earliest possible release date, and timing could slip further.
• OLED iMac: Apple is also working on an OLED iMac, though its release date is unknown. It could arrive around the same time as the OLED MacBook Air.

2026 Product Refresh Timeline

Apple has already released several products in 2026, including:

• January 2026 — AirTag 2
• March 2026 — iPhone 17e
• March 2026 — MacBook Neo
• March 2026 — M5 Pro and M5 Max MacBook Pro
• March 2026 — M5 MacBook Air
• March 2026 — M4 iPad Air
• March 2026 — Studio Display and Studio Display XDR
• March 2026 — AirPods Max 2
• June 2026 (Announced) — iOS 27, iPadOS 27, macOS 27 Golden Gate, visionOS 27, tvOS 27, watchOS 27

Previous Release Timelines for Context

Apple's launch history provides a useful reference. In 2025, the company released the iPhone 16e in February, new MacBook Air and Mac Studio models in March, and then the iPhone 17 series, Apple Watch Series 11, and AirPods Pro 3 in September. In 2024, Apple shipped the Vision Pro in February, updated iPad Pro and iPad Air in May, iPhone 16 models in September, and new Mac mini, MacBook Pro, and iMac models in October. These historical patterns help set expectations for future release windows.

Looking further back, Apple's 2023 lineup included the M2 Pro/Max MacBook Pro in January, the 15-inch MacBook Air in June, and the iPhone 15 series in September. The 2022 releases included the Mac Studio, iPhone 14 series, AirPods Pro 2, and Apple Watch Ultra. 2021 brought the M1 iPad Pro, AirTag, iPhone 13 series, and the redesigned 14-inch and 16-inch MacBook Pro.

Source: MacRumors News

Key Facts

• Starting price: $799
• Released: September 19, 2025
• Display: 1,245 sq mm LTPO3 OLED, 3,000 nits peak brightness
• Battery: up to 42 hours standard, 72 hours in Low Power Mode
• Connectivity: 5G, satellite, dual-frequency GPS, Bluetooth 5.3, UWB
• Durability: 49mm titanium, MIL-STD-810H, WR100/EN13319
• Chip: S10 with Double Tap and Wrist Flick gestures

Should You Buy the Apple Watch Ultra 3?

For outdoor enthusiasts who need extra durability and advanced tracking, the Apple Watch Ultra 3 is currently a solid buy. It was refreshed in September 2025, and with the next update expected around September 2026, there is plenty of time before a successor arrives. The improvements over the prior Ultra are meaningful, particularly the larger display, satellite communications, and improved battery life.

Compared to the standard Apple Watch Series 11, the Ultra 3 offers better water resistance, a more rugged enclosure, and significantly longer battery life. However, it is also larger at 49mm and may not suit everyone. Those with smaller wrists or those who do not need the extreme feature set should look at the Series 11 instead.

Design and Materials

Apple has retained the familiar look of the Apple Watch Ultra with a 49mm aerospace-grade titanium case. The material is lightweight yet durable, providing corrosion resistance for water-based activities. The third-generation model is available in Natural Titanium and Black Titanium, with the black version featuring a diamond-like carbon coating for improved scratch resistance.

The case is made using 3D printing technology with aerospace-grade titanium powder, reducing material waste compared to traditional manufacturing. The Digital Crown has larger, coarser grooves for use with gloves, and the side button is also glove-friendly. A customizable Action button sits on the left side in a bright orange color, allowing quick access to workouts, compass waypoints, and other functions.

Durability is a hallmark of the Ultra series. The watch is certified to MIL-STD-810H for military-grade environmental testing, including temperature extremes, shock, vibration, and immersion. It operates from -4 degrees Fahrenheit (-20 degrees Celsius) to 131 degrees Fahrenheit (55 degrees Celsius). The flat sapphire crystal display is more scratch-resistant than the glass used on standard models.

Water resistance is another key differentiator. The Apple Watch Ultra 3 is rated WR100 and EN13319, making it suitable for recreational scuba diving to 40 meters. It can also handle kitesurfing and wakeboarding, activities not recommended for standard Apple Watch models.

Display

The Ultra 3 features an LTPO3 wide-angle OLED display with smaller bezels, increasing the viewable area to 1,245 square millimeters. The display is larger than the 46mm Apple Watch Series 11 and retains up to 3,000 nits of peak brightness. It also maintains that brightness at wider viewing angles, making it easier to see outdoors.

The always-on display now supports a ticking seconds hand that is visible without raising the wrist. Brightness can drop to 1 nit in dark environments, which is useful in movie theaters or while trying not to disturb others at night.

S10 Chip and Gestures

Under the hood, the Apple Watch Ultra 3 uses the same S10 chip found in the Series 11. While there are no performance gains over the S9 in the previous Ultra, the chip supports new gesture controls introduced in watchOS 26. Double Tap allows users to answer calls, stop timers, play or pause music, and more by tapping their index finger and thumb together. Wrist Flick is a new gesture that dismisses notifications or declines calls with a simple flick of the wrist.

Storage remains at 64GB, providing plenty of space for apps, music, and offline data.

Health Features

The Apple Watch Ultra 3 includes a comprehensive suite of health monitoring tools. It uses a third-generation optical heart rate sensor for real-time heart rate data, an electrical sensor for ECG readings, and LEDs and infrared for blood oxygen measurements. The watch can detect high, low, and irregular heart rhythms, with notifications sent to the user when anomalies are detected.

One of the new features is Hypertension Alerts. The watch does not measure blood pressure directly but uses heart rate data over a 30-day period to identify potential signs of hypertension. Users who receive an alert can then perform blood pressure measurements with a cuff for seven days to share with their doctor.

Blood oxygen monitoring is available, although in the United States the data is sent to the iPhone and displayed in the Health app due to an ongoing patent issue. In other countries, readings appear directly on the watch.

Sleep Apnea Detection uses the accelerometer to analyze wrist movements during sleep for signs of interrupted breathing. After 30 days, the watch can indicate whether symptoms suggest sleep apnea. The Health app now includes a Sleep Disturbances category for this data.

Other health features include fall detection, crash detection, and temperature sensing. The Apple Watch Ultra 3 can alert emergency services if it detects a hard fall or a severe car crash. Temperature sensors support menstrual cycle tracking and retrospective ovulation estimates, as well as providing water temperature data in the Depth app.

Activity Features

Like all Apple Watches, the Ultra 3 tracks daily movement, workouts, and exercise. The large display shows the three Activity rings and provides reminders to stand and move.

For outdoor use, the watch includes a Depth app that displays current depth, water temperature, time underwater, and maximum depth. The Oceanic+ app, developed with Huish Outdoors, can turn the watch into a dive computer with a Bühlmann decompression algorithm.

A dedicated Tides app provides high and low tide times, tide height, and sunrise/sunset data. The Compass app includes a digital analog face, elevation, incline, latitude, and longitude. Users can drop Compass Waypoints to mark points of interest, and the Backtrack feature uses GPS to guide users back to their starting point if they become lost.

Battery Life

The Apple Watch Ultra 3 offers up to 42 hours of battery life on a single charge, a six-hour improvement over the Ultra 2. In Low Power Mode, battery life extends to 72 hours, making it suitable for multi-day trips. Fast charging delivers 80 percent capacity in 45 minutes, and a 15-minute charge provides up to 12 hours of normal use.

This battery performance is enough for users to complete a full long-course triathlon without needing to recharge partway through.

Connectivity

The Ultra 3 features built-in cellular connectivity with 5G support, a first for the Apple Watch. It also introduces satellite communications for off-the-grid situations. Emergency SOS via satellite, Messages via satellite, and Find My via satellite are available when Wi-Fi and cellular are unavailable.

The watch includes a second-generation Ultra Wideband chip for Precision Finding of a misplaced iPhone and compatibility with digital car keys. Bluetooth 5.3, Wi-Fi, and an 86-decibel siren with two alternating patterns round out the connectivity options. The siren can be heard up to 600 feet away.

Dual-frequency GPS with L1 and L5 bands provides the most accurate location tracking of any Apple Watch, which is essential for athletes who need precise distance, pace, and route data.

Comparison With Series 11 and SE 3

The Apple Watch Ultra 3 is sold alongside the Series 11 and the more affordable SE 3. The Series 11 starts at $399 and lacks satellite connectivity, the siren, and the rugged casing. The SE 3 starts at $249 and omits the blood oxygen sensor and ECG app. The Ultra 3 remains the top-end option for users who need the best battery life, durability, and advanced outdoor features.

What's Next for the Apple Watch Ultra

Apple is already working on an Apple Watch Ultra 4, but the next generation is expected to bring only incremental upgrades. A new S12 chip with faster performance and potential health tracking improvements are likely, while major design changes are not anticipated until at least 2028. Noninvasive blood glucose monitoring remains a future possibility, but it is still years away.

Apple may also adopt LG Display's high-mobility oxide technology for a future Apple Watch screen, which could reduce power consumption. Touch ID integration has been rumored for future models, hopefully providing new convenience features for unlocking and authentication.

The Apple Watch Ultra 3 continues to define the premium smartwatch category for outdoor enthusiasts, and future updates will focus on refining the experience rather than reinventing it.

Source: MacRumors News

Depuis plus de deux ans, la jeune femme, ex-influenceuse beauté et mannequin, voit ce point mauve la poursuivre comme une menace constante. Un ancien abonné, autrefois épris, s’est transformé en un serial harceleur, vouant son existence tout entière à la tourmenter. « Il m’envoie des menaces de viol tous les jours », raconte Marion en montrant ces messages de haine.

Les conséquences de cette entreprise de démolition s’avèrent terribles : dépression, pensées morbides, vie sociale en miettes, activité professionnelle au point mort. Sa plainte pour harcèlement et menaces de mort, au moment où nous la rencontrons, était au point mort dans un commissariat du nord de la France, laissant la jeune femme sans solution ou presque.

Une vie de mannequin

Dans sa vie d’avant, tout allait si bien pourtant. Marion n’est pas encore majeure quand elle se lance dans le mannequinat. Avec succès : les contrats s’enchaînent, elle gagne en visibilité via les réseaux sociaux (Instagram en tête), inévitables rampes de lancement. Elle y poste ses photos par dizaines. Parmi ses 30 000 abonnés, beaucoup d’hommes. Certains lui envoient des messages en privé, teintés de misogynie. Il y a aussi des insultes gratuites. Elle dit, comme une fatalité : « C’est presque normal dans notre métier. »

Cette expérience, elle la partage avec de nombreuses autres créatrices de contenu. Selon une étude menée en 2024 par l’institut Ifop pour l’association e-Enfance, 87 % des influenceuses ont déjà été confrontées à des messages à caractère sexuel non sollicités. Beaucoup ne portent pas plainte, par peur des représailles ou par découragement. Marion, elle, pensait que répondre calmement suffirait à désamorcer. « J’ai répondu à tout le monde au début. Je pensais que c’était le métier qui voulait ça. »

Un « soumis » devenu harceleur

Parmi les messages reçus, un abonné au profil singulier lui en écrit beaucoup. « Au début, il m’appelait « Ma reine » et me demandait de l’insulter », se souvient-elle. En langage d’influenceur, on appelle cela un « soumis ». Puis il a commencé à lui raconter sa vie : pas d’amis, une famille compliquée, une profonde solitude. Marion n’ose pas le bloquer. « Je lui répondais toujours, par humanité. Il me faisait presque de la peine. Si j’avais su… », regrette-t-elle.

Laure Salmona, directrice de l’association Féministes contre le cyberharcèlement, analyse ce mécanisme : « Une illusion de proximité se crée. L’abonné interprète la moindre réponse comme une attention personnelle. Son obsession se renforce à mesure que l’influenceuse se montre bienveillante. » Elle ajoute que les personnes isolées sont plus susceptibles de développer ce type de comportement, sans qu’il soit toujours possible de le prédire.

À l’été 2024, Marion souhaite couper les ponts. Elle lui signifie par message. La réponse fera tout basculer : « Moi et toi, c’est pour la vie », prévient l’internaute, qui agit sous pseudonyme. Il tente alors de la faire chanter psychologiquement. Ne parvenant pas à la faire céder, il passe à la vitesse supérieure.

Une intimidation sans limite

Le harcèlement prend une dimension méthodique et cruelle. Il crée des dizaines de comptes TikTok et Instagram à l’effigie de Marion, accompagnés d’insultes : « Marion la tana » (une insulte équivalente de « salope »). Il détourne des photos de sa mère, affublée d’un nez de cochon, et fait de même avec son père, ses cousines et ses neveux. « Il a menacé de les violer par message. Ils ont 10 ans à peine, je ne sais pas si on se rend compte ? », s’indigne-t-elle.

Les marques qui collaboraient avec Marion sont également ciblées. L’harceleur les contacte, leur envoie des messages diffamatoires, commente sous leurs publications pour les dissuader de poursuivre leurs contrats. L’une après l’autre, les enseignantes retirent leurs offres. « Je

Source: leparisien.fr News

L'information est d'autant plus importante que Vin Diesel a également évoqué la possibilité de faire revenir Brian O'Conner, le personnage rendu célèbre par Paul Walker. Une idée qui semble impossible à première vue, puisque l'acteur est décédé le 30 novembre 2013 dans un accident de voiture. Pourtant, les progrès des effets visuels ont déjà permis à la production de ressusciter son image à l'écran dans le septième volet. La technologie a considérablement évolué depuis, et ce retour pourrait être cette fois plus réaliste que jamais.

Une annonce faite au coeur de la culture automobile

Le Fuel Fest, qui se tenait en Californie, réunit chaque année des milliers d'amateurs de voitures, de tuning et de courses de rue. C'est exactement l'univers qui a vu naître la franchise en 2001, lorsque le premier film présentait au public un groupe de jeunes passionnés qui organisaient des courses illégales dans les rues de Los Angeles. En choisissant cet événement pour parler de l'avenir de la série, Vin Diesel a clairement montré sa volonté de revenir à l'esprit originel.

L'acteur était accompagné de deux autres figures importantes de la saga : Tyrese Gibson, qui incarne Roman Pearce depuis le deuxième film, et Cody Walker, le frère cadet de Paul Walker. La présence de Cody Walker était particulièrement remarquée, car il a participé à des reshoots de Fast and Furious 7 après la mort de son frère. Elle a naturellement alimenté les spéculations sur un éventuel retour de Brian O'Conner. Les fans présents sur place ont tout de suite compris que l'annonce serait d'une importance capitale.

Les trois conditions imposées par Vin Diesel

Selon les propos rapportés depuis l'événement, Vin Diesel a raconté que le studio Universal lui avait demandé de préparer une sortie pour avril 2027. Le comédien aurait répondu qu'il acceptait, mais à trois conditions. La première est de ramener toute la franchise à Los Angeles, la ville originelle de la saga. La deuxième consiste à revenir à la culture automobile et aux courses de rue, qui avaient peu à peu laissé place à des missions d'espionnage et à des cascades de plus en plus spectaculaires. La troisième condition, sans doute la plus émouvante, est de réunir Dom et Brian O'Conner sur un même écran une dernière fois.

Ces trois exigences peuvent se lire comme un manifeste. Vin Diesel a souvent répété que la famille était au coeur de Fast and Furious. En remettant la course automobile au premier plan, il s'agit de retrouver l'essence même de la franchise. En alternant avec le personnage de Paul Walker, il s'agit d'offrir une conclusion émotionnellement forte à une histoire qui a commencé avec deux pilotes de rue.

Brian O'Conner, un personnage emblématique

Brian O'Conner était au départ un policier infiltré dans le milieu des courses de rue. Sa rencontre avec Dominic Toretto a transformé le premier film en une histoire d'amitié et de loyauté. Au fil des épisodes, il est devenu un membre à part entière de la famille Toretto. Le personnage a été interprété par Paul Walker dans la majorité des films de la saga, et il est resté dans le coeur des spectateurs bien après sa disparition. Son absence dans les derniers volets a été ressentie comme un vide immense.

Faire revivre Brian O'Conner nécessitera un travail de préparation considérable. La production dispose d'images de Paul Walker issues des précédents films, mais aussi de scènes inédites tournées avant sa mort. Grâce à des techniques de modélisation 3D, il sera possible de créer un personnage numérique fidèle à l'acteur. Les équipes techniques devront également reconstituer sa voix, soit en utilisant des enregistrements existants, soit en travaillant avec des comédiens capables de l'imiter. Le résultat final devra respecter la mémoire de l'acteur et son héritage.

La mort de Paul Walker, un choc planétaire

La disparition de Paul Walker a bouleversé le tournage de Fast and Furious 7. Le comédien est mort dans un accident de Porsche, à Santa Clarita, alors qu'il participait à une oeuvre caritative. Il avait 40 ans. Son absence a obligé les scénaristes à réécrire la fin du film, prévue initialement comme une suite. Les frères de Paul, Caleb et Cody, ont accepté de se rendre sur le plateau pour servir de doublures. Le visage de l'acteur a ensuite été appliqué numériquement sur leurs corps dans plusieurs scènes.

Ce drame a profondément affecté Vin Diesel, qui considérait Paul Walker comme un frère. La relation entre leurs deux familles s'est renforcée au fil des années. La fille de Paul Walker, Meadow, est régulièrement apparue aux côtés de Vin Diesel lors de grandes cérémonies. Il a notamment accompagné Meadow Walker jusqu'à l'autel lors de son mariage, un geste chargé d'émotion. Cette attention a ému les fans et confirmé que les liens créés par la franchise dépassaient

Source: CNEWS News

A Homecoming at Midwest Gymnastics Center

Monday marked Lee's first official day back at Midwest Gymnastics Center as she begins preparing for what she hopes will be her third Olympic Games. The decision about where to train, Lee said, was never complicated. The gym is not just a training facility for her; it is a sanctuary, a second home, and the place where the people who know her best have always believed in her.

“Allie and Jess are like my parents,” Lee said, referring to Alison and Jess Graba, who have been pillars of her life since she was a child. Jess Graba has coached Lee since she was 6 years old, when she was a tiny, energetic kid who loved flipping around her house. Alongside his wife, Alison, he helped transform that raw enthusiasm into one of the most decorated gymnastics careers in American history.

“We've been through so much together,” Lee said. “I honestly couldn't see myself with anybody else, and I wouldn't want to do this process with anybody else, especially for L.A.”

A Trusted Team, Now Reunited

This Olympic cycle also brings another familiar face into the fold. Jeff Graba, Jess's identical twin brother and Lee's former college coach at Auburn University, has rejoined the coaching staff. Jeff's return gives Lee a trusted team that has been part of nearly every chapter of her gymnastics career. The brothers have spent years coaching together, joking together, and occasionally confusing everyone else with their identical looks. For Lee, having both of them by her side is more than just a comfort; it is an essential part of her process.

“They know me better than I know myself,” Lee said. “They understand me and how I work as an athlete, but also just as a human.” That trust, forged over years of wins, losses, injuries, and comebacks, has become one of Lee's biggest reasons for returning home. She knows that the road to a third Olympics will be grueling, and she wants to travel it with people who see her not just as a medal count, but as a person.

Jess Graba echoed that sentiment. “It's just too much work to do something like this and not have fun,” he said. “You better enjoy doing it and who you're doing it with.” The joy of the sport, he added, is central to Lee's longevity. She has never been the kind of athlete who trains purely for external validation; she trains because she loves the challenge, the process, and the community around her.

Reflection and Rediscovery After Paris

Following the Paris Olympics, where Lee helped the U.S. women's gymnastics team win gold and added individual bronze medals in the all-around and uneven bars, Lee stepped away from the competitive spotlight. She spent time living in New York, building her personal brand, and reflecting on what came next. That period was essential, she said, for helping her regain confidence and rediscover herself outside of gymnastics. The time away gave her clarity about what she wanted for her future, and ultimately, it led her back to the sport she has loved since childhood.

But the road back was not without its challenges. Lee has been open about her battle with kidney issues, which she publicly revealed in 2023. She has spoken candidly about the physical and mental toll that competing at the highest level takes, especially while managing a chronic condition. Her resilience in the face of these obstacles has made her a role model for athletes and non-athletes alike. The decision to train for another Olympic cycle was not made lightly, but Lee says she feels healthier and more grounded than she has in years.

The Minnesota gym where she trains has become more than just a place to practice routines. It is a gathering place where young gymnasts look up to her, where families cheer for her, and where Lee says she feels completely at home. “I love getting to come back and seeing all of the little kids and all of them running up to me,” she said. That connection to the community has always been part of her motivation. She remembers being that little kid, dreaming of the Olympics, and she carries that memory with her every time she steps onto the floor.

A Legacy of Excellence

Lee's gymnastics resume is remarkable by any standard. At the Tokyo 2020 Olympics, held in 2021 due to the pandemic, she won the all-around gold medal, becoming the first Hmong American to compete in the Olympics and the first Asian American woman to win the all-around title. She also took home a team silver and a bronze on the uneven bars in Tokyo. Four years later in Paris, she proved her staying power by helping the U.S. team reclaim gold and adding individual bronze medals in the all-around and uneven bars. Her performances have made her one of the most recognizable faces in American gymnastics.

But her journey has never been just about the medals. Growing up in Saint Paul, Minnesota, Lee was part of a large, close-knit Hmong family. Her father, John Lee, was a crucial influence in her early years, building a balance beam in their backyard and encouraging her to pursue her dreams. When her father suffered an accident that left him paralyzed, Lee has said that she competed in Tokyo with him in mind, and his strength pushed her through the pressure of the Olympic stage.

She also made history as a collegiate athlete at Auburn University, where she balanced NCAA competition with elite gymnastics, a feat that is extremely demanding. Her college career further solidified her bond with Jeff Graba, who coached her alongside the rest of the Auburn staff. The connections she built there have stayed with her, and they are now part of the reason she feels confident about the next chapter.

Preparing for a Third Olympic Run

The road to the 2028 Los Angeles Olympics is a long one, and Lee knows that she will need to manage her body and her schedule carefully. The U.S. Olympic Gymnastics Trials are scheduled for June 2028 in Louisville, Kentucky. The competition will determine who earns a spot on Team USA for the Los Angeles Games, where gymnastics will once again be one of the marquee events. Lee will be in her late twenties by then, which is considered older for an elite gymnast, but she has already shown that she can defy expectations.

Her training at Midwest Gymnastics Center will focus on rebuilding strength, refining technique, and introducing new or upgraded skills to keep her competitive against a deep field of American gymnasts. The United States has an abundance of talent, with young gymnasts rising through the ranks and veterans like Simone Biles also contemplating their futures. Lee's experience, artistry, and consistency make her a valuable asset to any team, and she is determined to be in the conversation when the Olympic trials arrive.

The support of the Graba family has been a constant throughout her career. Jess and Alison have coached Lee through injuries, losses, and moments of doubt. They have celebrated her victories and helped her navigate the pressures of fame. In many ways, they have become her second family, and their presence is a reminder of the small, humble beginnings that launched her into the global spotlight.

“I think it's been fantastic, honestly,” Jess Graba said of the reunion with his brother Jeff and the return of Lee. “It's been a really good full-circle moment.” Indeed, the scene at Midwest Gymnastics Center now resembles the early days when Lee first dreamed of making it to the Olympics. The same coaches, the same gym, the same community, and the same passion are all there.

For Lee, the journey is far from over. She is embracing the process, surrounded by the people who believed in her long before the rest of the world did. As she begins this new chapter, her message is simple: she wants to enjoy the ride, push her limits, and see how far she can go with the family she found at the gym. The Los Angeles Games may be more than a thousand days away, but Lee's focus is already squarely on the opportunity ahead. She knows that the work starts now, and she is exactly where she wants to be.

Source: MSN News

In 2026, the market is full of fast chargers across every price bracket. Some are small enough to slip into a pocket, while others are designed as desktop stations with enough power to charge an entire technology kit. The best models use gallium nitride (GaN) technology, support USB Power Delivery, and offer a sensible mix of USB-C and USB-A ports. Whether you're replacing a lost adapter or upgrading your setup, here is everything you need to know about fast chargers in 2026.

Key facts at a glance

• Fast chargers are essential because many phones, tablets, and laptops no longer ship with power bricks.
• GaN technology makes chargers smaller, lighter, and cooler than older silicon-based adapters.
• USB Power Delivery is the most common open fast-charging standard, but proprietary standards like SuperVOOC can be faster for specific phones.
• Multi-port chargers share a fixed total wattage, so power distribution matters when charging multiple devices at once.
• For most smartphones, any charger above 15W to 18W is considered fast; high-end laptops can draw up to 240W.

Best fast chargers for 2026

After testing dozens of adapters for speed, build quality, port selection, and value, the following categories cover the standouts for 2026:

• Best overall fast charger: A 100W GaN wall charger with two USB-C ports and one USB-A port is the sweet spot for most users. It can fast-charge a laptop and a phone simultaneously while still being compact enough for a bag.
• Best compact charger: A 30W to 45W single-port GaN charger is small enough for a pocket but fast enough for a phone or a small tablet. This is the ideal replacement for the charger that no longer comes in the box.
• Best travel charger: A 65W foldable-plug charger with multiple ports and international voltage support is ideal for crossing borders. Look for models with interchangeable plugs or universal voltage input.
• Best desktop charger: A 140W to 200W multi-port station can power a laptop, tablet, phone, and earbuds at once. Desktop chargers are larger, but they provide the highest total power and the most ports.

These categories represent the three main use cases: minimalists who charge one device at a time, road warriors who need a portable brick, and power users who need a desktop hub. Brands such as Anker continue to lead the shift toward GaN, offering chargers that are dramatically smaller than the silicon bricks of the past.

What to consider before buying a fast charger

Before you start shopping, ask yourself three questions: How many devices do you need to charge? How much power do they require? Will you be traveling with the charger? The answers will narrow your options considerably.

If you only charge a single phone, a low-wattage single-port charger is cheaper and more compact than a multi-port brick. If you carry a laptop and a phone, a 65W or 100W two-port charger can handle both. If you are managing a full kit of devices, a desktop charger with three or more ports and smart power sharing is worth the extra desk space.

Power requirements

Every device has a maximum charging speed, usually expressed in watts. A smartphone typically accepts 20W to 45W, although some high-end Android models can go beyond 100W with a proprietary charger. A tablet usually needs 20W to 45W as well. A standard ultrabook or MacBook Air-style laptop can charge with a 30W to 60W adapter, while a 14-inch or 16-inch pro laptop may need 100W or more.

Gaming laptops are a different story. Some of these machines draw more than 140W, and they often use proprietary barrel connectors. They may not be compatible with universal USB-C chargers at full power. Even if a USB-C charger can provide some energy, it will charge at a slower rate than the laptop's original adapter. Check your device's spec sheet before buying a universal brick.

Port layout and power sharing

Port layout matters more than ever. A charger with two USB-C ports and one USB-A port covers the most ground: you can charge a laptop, a phone, and an older accessory at the same time. However, the total wattage is shared among all active ports. A 100W charger might output 65W through the first USB-C port, but if you plug in a second device, the power is divided, often to 60W and 40W, or 45W and 45W depending on the model. Check the charger's spec sheet to see how it distributes power across each port combination.

Cables also play a role. A cheap USB-C cable rated for only 60W can bottleneck a 100W charger. For full-speed charging, you need a cable rated for the same power level. The USB Power Delivery 3.1 specification supports up to 240W, and high-quality cables are clearly labeled with their power rating. If you are using a Lightning cable, you will need a USB-A port or a cable with the appropriate connector type.

Travel-friendly features

Frequent travelers should look for foldable prongs, multi-voltage input (100-240V), and compact dimensions. A GaN charger with foldable pins is much easier to slip into a backpack or carry-on. Some chargers also come with interchangeable international plugs, which can be a huge convenience if you fly between regions. If you're already traveling with power banks or a charging station, choosing a compact GaN charger helps streamline your gear.

Why GaN is everywhere

Gallium nitride, or GaN, is a semiconductor material that has transformed charger design. Compared with older silicon-based adapters, GaN chargers are more efficient, generate less heat, and can be 30 to 50 percent smaller and lighter for the same power output. A 65W GaN charger can be barely larger than a traditional 20W brick. This makes GaN particularly valuable for travelers and for anyone who carries a charger in a pocket.

The efficiency of GaN also enables higher power levels in compact enclosures. A 140W GaN charger can fit in the palm of your hand, while a silicon-based equivalent might be twice the size. As a result, GaN has become the standard for premium chargers in 2026.

Charging standards explained

Not all fast charging is the same. The most common open standard is USB Power Delivery (USB PD), which allows the charger and the device to negotiate voltage and current. Many modern chargers also support PPS (Programmable Power Supply), which adjusts voltage in small increments to match a phone's ideal charging curve. PPS can reduce heat and improve charging speed on compatible devices.

Qualcomm Quick Charge is another fast-charging technology, often found on Android phones. It is less common in laptops and tablets. However, most modern phones and chargers support USB PD at a baseline, which means they can charge at decent speeds even if Quick Charge or a proprietary standard is not available.

Proprietary standards can be significantly faster for specific phones. OnePlus, for example, uses SuperVOOC; its included brick can deliver up to 100W to a OnePlus 13, but a generic USB-PD charger will max out at 45W. That is still fast, but it shows why it is important to know which standard your phone uses before buying an adapter.

Fast charger FAQs

What is GaN?

Gallium nitride is a compound semiconductor that conducts electricity more efficiently than silicon. In chargers, it allows for higher switching frequencies, which means smaller transformers and capacitors are needed. The result is an adapter that is much smaller than a traditional silicon-based charger of the same wattage. GaN chargers also run cooler and support higher power levels, making them the standard for premium fast chargers in 2026.

Do fast chargers affect battery life?

Fast charging does generate more heat than slow charging, and heat is a known contributor to long-term battery degradation. However, modern devices and chargers use multiple safeguards to keep temperature and voltage within safe limits. Charging protocols automatically slow down as the battery approaches full, and many phones stop charging briefly if the temperature rises too high. As long as you use a compatible charger and a good cable, the effect of fast charging on overall battery health is small.

What is the difference between a fast charger and a regular charger?

There is no universal definition of fast charging, but a common benchmark is any charger that delivers more than 15W to 18W. Older chargers often delivered just 5W, which is painfully slow for a modern smartphone. At the high end, phones can receive more than 100W, and laptops can draw up to 240W under the USB Power Delivery 3.1 standard. The term 'fast' is relative to your device's needs and capabilities.

Matching a charger to your device

When comparing chargers, focus on the device you own, not the highest number on the box. A 140W charger will not charge a 20W phone any faster than a 20W charger; the phone's charging circuit controls the amount of power it can accept. Buying a charger with a bit more power than you need gives you headroom for future devices, but it is not necessary for a single phone.

For a modern iPhone, a 20W USB-C charger is enough to take advantage of fast charging. For an Android flagship, a 45W charger will cover most models, and a 100W charger provides extra room. For a laptop, choose a charger that matches the laptop's original adapter wattage, or at least meets the minimum recommended power. A 65W charger can handle many ultrabooks, while a 100W charger is safer for larger laptops. If you frequently charge multiple devices at once, a 100W or 140W multi-port GaN charger is the safest bet.

Travelers should also remember that AirPods, Apple Watch, and other small wearables have low power requirements. They can share a port or use a lower-wattage output without issue

Source: Engadget News

Finding yourself far from a wall socket when your phone hits five percent is positively nerve-wracking. If you stash a portable battery in your bag, you can avoid that feeling altogether. But there are thousands of power banks out there and it can be tough to pick the right one for what you need. We've spent years testing dozens of batteries and found the best power banks for different scenarios. Whether you need a quick re-up for your phone or a huge brick to keep your laptop alive, you'll find something fitting here.

Best MagSafe power bank

Anker MagGo Power Bank (10K)

Capacity: 10,000mAh | Maximum Output: 15W (wireless) | Ports: One USB-C in/out | Included cable: USB-C to USB-C | Charges iPhone 15: 1.64 times | Charge time iPhone: 4 to 100% in 2h 26m | Weight: 8.82 oz | Dimensions: 4.22 x 2.71 x 0.78 in

The Anker MagGo Power Bank was one of the first Qi2-certified products on the market, and the new standard has made the popular MagSafe/kickstand model much faster. It's the most well-rounded MagSafe battery we've tested. It brought an iPhone 15 from near-dead to half-full in about 45 minutes, compared to an hour and a half for previous top picks. After that initial refill, the MagGo 10K had enough left over to get the phone to 70 percent on a subsequent charge.

This wireless power bank adds an LCD display to show remaining battery percentage and charge time estimates. A strong MagSafe connection lets you use the phone while it charges, and the small kickstand creates a sturdy base for watching videos. New Pixel 10 phones with Qi2 support also benefit from faster wireless charging and magnetic alignment. The USB-C port is there for non-wireless devices.

• Pros: Qi2 fast wireless charging, sturdy kickstand, LED display
• Cons: More expensive than other MagSafe packs

Best portable charger for Android

Anker Nano Battery (Foldable USB-C)

Capacity: 5,000mAh | Maximum Output: 22.5W | Ports: One USB-C and one USB-C connector | Charges Galaxy S23 Ultra: 0.65 times | Charge time: 0 to 65% in 1h 2m | Weight: 3.5 oz | Dimensions: 3.03 x 1.45 x 0.98 in

The Anker Nano power bank delivers impressive power for its size. It's the exact shape of a lipstick case and has a built-in USB-C connector that folds down when not in use. That means you don't need to remember a cable. A built-in USB-C port can refill the battery or charge another device with an adapter. Four indicator lights show the remaining charge.

In testing, the 5,000mAh battery got a depleted Galaxy S23 Ultra back to 65 percent in about an hour. It's small enough to use your phone while charging without feeling awkward, and it's also a great pick for recharging earbuds. For more juice, Anker's 30W Nano Power Bank offers 10,000mAh, a display, and a built-in cable that doubles as a carry handle.

• Pros: Very portable, affordable
• Cons: Small enough to get misplaced

Best low-capacity power bank

Anker 10K Fusion

Capacity: 10,000mAh | Maximum Output: 30W | Ports: One USB-C in/out port, one USB-C in/out cable, wall prongs | Charges iPhone 15: 1.86 times | Charge time iPhone: 5 to 100% in 1h 53m | Charges Galaxy S23 Ultra: 1.45 times | Charge time Galaxy: 5 to 100% in 1h 2m | Weight: 8.8 oz | Dimensions: 4.25 x 2.0 x 1.22 in

The toughest thing about using a power bank is remembering to bring it, along with a cable and wall adapter. Anker's 10K Fusion solves those problems with its attached USB-C cable and foldable wall prongs. It's compact, just a smidge wider than a stick of butter, yet packs 10,000mAh capacity. The 30W output enabled Super Fast Charging on a Galaxy S23, taking it from 5 percent to full in just over an hour. In 20 minutes, it bumped a near-dead iPhone 15 to 45 percent.

There's an additional USB-C port for other cables, and both the built-in connector and the port can refill the power bank. The display shows remaining charge accurately, and the corduroy side texture is a nice touch. The previous pick in this category, the BioLite Charge 40 PD, is still excellent, but the 10K Fusion delivers faster charging and more features at a lower price.

• Pros: Built-in USB-C cable, built-in wall prongs, accurate display, affordable
• Cons: iPhone charging is slower than some rivals

Best medium capacity power bank

Belkin BoostCharge 20K with Integrated Cable — $50

Capacity: 20,000mAh | Maximum Output: 30W | Ports: One built-in USB-C in/out cable, one USB-A port, one USB-C port | Charges iPhone 15: 3 to 3.5 times | Charge time iPhone: 5 to 100% in 2h 6m | Charges Galaxy S23 Ultra: 2.5 to 3 times | Charge time S23 Ultra: 1h 15m | Weight: 14 oz | Dimensions: 6.06 x 3.0 x 0.99 in

An integrated cable is a hot new feature in portable chargers, and the Belkin BoostCharge 20K is a prime example. It outputs up to 30W, which isn't the fastest, but it charged a Galaxy S24 Ultra from dead to full in 75 minutes and an iPhone 15 from 5 to 87 percent in just over an hour. The 20,000mAh capacity delivers those numbers about three times over.

The built-in USB-C cable is conveniently magnetized, and there are two additional ports: an out-only USB-A and an in/out USB-C. You can charge three devices simultaneously, though speeds will drop. Instead of a digital screen, it has four accurate LED indicator lights. The BoostCharge comes in blue, pink, white, or black, and the matte finish repels smudges well.

• Pros: Built-in cable, four color options, affordable, great capacity for the price
• Cons: Not the fastest charging

Best medium-high capacity power bank

Nimble Champ Pro — $100

Capacity: 20,000mAh | Maximum Output: 65W | Ports: Two USB-C in/out | Charges iPhone 11: 2.95 times | Charge time iPhone: 5 to 100% in 1h 39m average | Charges Galaxy S22 Ultra: 2.99 times | Charge time Galaxy: 5 to 100% in 59m average | Charges iPad Air: 1.83 times | Weight: 12.9 oz | Dimensions: 5.92 x 2.48 x 1.00 in

Nimble's Champ Pro delivers a screaming fast charge. It got a Galaxy S23 Ultra from 5 percent to full in under an hour, faster than every other battery we tested except for the premium Anker Laptop Power Bank. It provides nearly three full charges to an iPhone or Galaxy device and can refill an iPad more than once. The battery itself recharges from the wall noticeably faster than other models.

Nimble is a certified B-Corp that uses 90 percent post-consumer recycled plastic and eco-friendly packaging. The unit feels sturdy, with a compact shape and an adjustable lanyard. Both USB-C ports are in/out, letting you charge two devices at once. The only downside: the four indicator lights underestimated the remaining charge in early testing, though accuracy improved over time.

• Pros: Super fast charging, made from recycled materials, sturdy compact design
• Cons: Indicator lights underestimate charge

Best multi-device power bank

BioLite Charge 100 Max

Capacity: 25,000mAh | Maximum Total Output: 120W | Ports: Wireless pad (15W), two USB-C (100W), one USB-A (15W), one USB-C (15W) | Charges iPhone 15: 5 times | Charges Galaxy S23 Ultra: 4 times | Charges iPad Air: 2.2 times | Charges MacBook Pro: 0.75 times | Weight: 1.28 lbs | Dimensions: 5.5 x 4.4 x 1.38 in

The BioLite Charge 100 Max has a compact, rounded design that's great for travel. It has four USB ports (three USB-C and one USB-A) plus a MagSafe-compatible wireless charging pad with 15W output. The magnetic hold is strong enough to keep it in place, though not for carrying around. The 10 LED pips are accurate, and the rubberized texture with yellow accents is a welcome change from typical black bricks.

The bank delivers quick charges to phones, tablets, and laptops. BioLite is a climate-neutral-certified B-Corp that helps bring energy to underserved areas. It costs more than some similar-capacity competitors, but the design and brand mission justify the premium.

• Pros: Compact colorful design, fast charging for multiple devices, climate-neutral company
• Cons: More expensive than similar-capacity batteries

Best laptop power bank

Anker Laptop Power Bank with Built-in Cable

Capacity: 25,000mAh | Maximum Output: 165W | Ports: Two built-in USB-C in/out cables, one USB-A port, one USB-C port | Charges iPhone 15: 4 to 5 times | Charges Galaxy S23 Ultra: 3.75 to 4 times | Charges iPad Air: 1.75 to 2 times | Charges MacBook Pro: 0.68 times | Weight: 1.31 lbs | Dimensions: 6.18 x 2.12 x 1.93 in

The Anker laptop power bank makes sure you're never without a way to charge. It has two attached USB-C cables: one on the side that acts as a carrying handle and a retractable cord that extends up to two feet. Both are in/out, so they can power devices or recharge the bank itself. The display shows remaining charge, output wattage per port, and estimated time to full when refilling.

It's an attractive, durable unit with a matte silver exterior. The display is smaller than the previous Prime bank, which was prone to scratches. At just $15 more than Anker's popular PowerCore bank, the built-in cables make it well worth the extra cost.

• Pros: Two built-in cables, durable build, detailed display, fast charging
• Cons: Screen picks up smudges

Best premium power bank

Anker Prime Power Bank 26K 300W

Capacity: 26,250mAh | Maximum Combined Output: 300W | Ports: Two USB-C (140W), one USB-A (22.5W) | Charges iPhone 15: 5 to 5.5 times | Charges Galaxy S23 Ultra: 4.3 times | Charges iPad Air: 2.5 times | Charges MacBook Pro: 0.83 times | Weight: 1.32 lbs | Dimensions: 6.3 x 1.5 x 2.5 in

The Anker Prime Power Bank stands out with its app connectivity, but the real stars are the speeds and capacity. The two USB-C ports can deliver up to 140W each, enough to charge two high-powered laptops simultaneously. It got a near-dead iPhone 15 to 60 percent in 30 minutes and delivered more charge to a MacBook Pro than any other battery we've tested.

The display shows remaining charge and battery temperature, a wise feature for lithium-ion packs. The sleek design has a shiny black front and matte silver body, and it's more compact than most 27,000mAh batteries. An optional charging base ($110) makes recharging effortless—just set the battery down and walk away. It's expensive, but for a truly premium power bank, this is it.

• Pros: Super fast charging, sleek premium design, display with temperature gauge
• Cons: Pricey, especially with optional base

Best power bank for outdoors

Nestout 15000mAh Outdoor Battery Power Bank

Capacity: 15,000mAh | Maximum Output: 32W | Ports: One USB-C in/out, one USB-C in, one USB-A | Charges iPhone 11: 2.99 times | Charges iPad Air: 1.17 times | Weight: 12.8 oz | Dimensions: 5.0 x 1.25 x 3.0 in

The Nestout Portable Charger has an IP67 rating, meaning it can survive submersion in a meter of water for 30 minutes. We tested by dropping it in a five-gallon bucket for five minutes, and it performed perfectly after drying off. The water resistance comes from screw-on caps with silicone gaskets. It also survived repeated drops from chest height onto hard surfaces.

Charging speeds aren't the fastest, but they're adequate. The included USB-A to USB-C cable is just seven inches long, so you'll likely want your own cord. Nestout offers clever accessories, including a snap-on LED worklight, a small tripod, and a portable solar panel that refilled the battery to 40 percent in under three hours.

• Pros: Waterproof with caps secured, clever accessories, survived drop tests
• Cons: Not the fastest, short included cable

What to look for in a portable battery pack

Battery type

Nearly every rechargeable power bank contains a lithium-ion battery. These beat other battery types in size-to-charge capacity and have increased in energy density by eightfold over the past 14 years. They also don't suffer from the memory effect that plagued older nickel-based batteries.

Flying with portable batteries

Lithium-ion batteries can overheat and catch fire, so there are restrictions on air travel. Current TSA rules allow external batteries rated at 100Wh or less (all our recommendations are) in carry-on luggage only—never in checked bags. Some airlines, like Southwest, now require power banks to be kept in clear view when actively charging a device.

Capacity

Capacity is listed in milliamp hours (mAh). A 5,000mAh battery can fill a smartphone to between 50 and 75 percent. Larger 20,000mAh and 25,000mAh packs can recharge laptops or give phones multiple charges. Note that efficiency is around 60 percent due to heat loss and voltage conversion. A 10,000mAh bank typically delivers about 5,800mAh to a device, while 20,000mAh delivers around 11,250mAh.

Wireless

Wireless charging is less efficient than wired, but it's convenient. Qi2-certified batteries support speeds up to 15W and include magnetic attachment points. MagSafe iPhones were the first Qi2-compatible devices, and now Google's Pixel 10 phones support the standard. The new Qi2 25W standard is supported by iPhone 17 and Pixel 10 Pro XL, and Qi2-25W power banks are starting to appear.

Ports

USB-C ports deliver faster charges than USB-A. Most modern power banks have Type-C connections, but Type-A jacks are still useful for older cables. Larger banks may list different wattages per port. The port's wattage determines what you can charge; a 15W port won't do much for a laptop. Cables must also match the maximum wattage. Some of the best power banks include built-in cables, so you're never without the right cord.

How we test power banks

We start with trusted brands and customer reviews, then acquire promising candidates for hands-on testing. For each battery, we charge an iPhone and an Android phone, plus an iPad and a MacBook Pro for larger units. Devices are drained to between 0 and 5 percent, then charged until full or until the power bank dies. We note charge times, the number of charges delivered, and the battery's efficiency. We continuously update this guide as new products are released.

Other power banks we tested

Belkin Stage PowerGrip: A 9,300mAh power bank with wireless charging, a built-in cable, Bluetooth shutter, and tripod thread. It's clever for photographers but slow, taking two hours to charge an iPhone 16 to 98 percent.

Anker MagGo for Apple Watch Power Bank: A 10K battery with a built-in USB-C cable and a pop-up Apple Watch charger. It's niche but saved us on trips when the watch was dying.

HyperJuice 245W: A sleek 27,000mAh brick with four USB-C ports. It's fast and powerful but lacks a USB-A port, and it's more expensive than comparable options with built-in cables.

EcoFlow Rapid Magnetic Power Bank: A Qi2-enabled 5,000mAh charger with a pull-out stand and attached cable. It looks nice but didn't outperform the top MagSafe pick.

Mophie Snap+ Powerstation Mini: A well-built 5,000mAh MagSafe battery with a sturdy stand, but only a partial charge for most phones. The current MagSafe pick offers double the capacity and a display for just $20 more.

Power bank FAQs

What's the difference between a portable power bank and a portable charger?

The terms portable battery, portable charger, external battery pack, and USB charger are used interchangeably. They all mean a lithium-ion battery that stores a charge to refill devices. Power stations, on the other hand, are much larger units—often the size of a car battery—that can't be taken on airplanes.

Does fast charging ruin your battery?

Not exactly. The real enemy is heat. Fast charging generates more heat, but modern phones have heat shields, heat sinks, and software to manage temperatures. Studies on EV batteries show a slight capacity decrease with fast charging, but ambient heat matters more. In short, fast charging is slightly harder on a battery, but built-in safeguards make the difference negligible.

Can you use a power bank for all your devices?

It depends on the bank's capacity and your device's battery size. A small 5,000mAh battery won't charge a laptop, but a 20,000mAh pack will give your computer a partial refill. Port compatibility matters too. If your device uses a DC or AC port, you won't be able to use a standard USB power bank.

Source: Engadget News

Best E Ink tablets for 2026

reMarkable 2 — Best E Ink tablet for most people

Screen size: 10.3-inch | Battery life: Two weeks | Network connectivity: Wi-Fi | Capacity: 8GB | Included stylus: No, extra cost | Supported file types: PDF, EPUB

The reMarkable 2 is our top pick because it offers the best mix of features for most users. It's lightweight, sleek, and provides an excellent writing and reading experience. The low-latency stylus feels remarkably close to pen-and-paper, and the device supports PDFs and EPUBs, making it great for academic papers and work documents. You can transfer files via desktop, mobile app, or cloud integrations with Google Drive, Microsoft OneDrive, and Dropbox. The optional Connect subscription ($3/month or $30/year) adds unlimited cloud storage and syncing, but the tablet is fully functional without it. At $399 (plus the cost of a marker), it's not cheap, but it's a solid investment for dedicated note-takers.

Pros: Great reading and writing experience; Google Drive, OneDrive, and Dropbox support; easy to use.

Cons: Marker costs extra; expensive; unlimited cloud storage requires a subscription.

reMarkable Paper Pro — Best premium E Ink tablet

Screen size: 11.8-inch | Battery life: Two weeks | Network connectivity: Wi-Fi | Capacity: 64GB | Included stylus: No, extra cost | Supported file types: PDF, EPUB

The reMarkable Paper Pro takes everything we loved about the reMarkable 2 and adds a larger color display (using a modified E Ink Gallery 3 technology called Canvas Color), better performance, and a more refined design. The 11.8-inch screen is great for writing, and the color capability is a welcome addition for highlighting and organizing notes. With a faster processor and lower pen latency (12ms), the Paper Pro feels zippy and responsive. It's expensive at $629, but if you want a premium E Ink experience, this is the one to get.

Pros: Color is a useful addition; backlight for dark environments; improved performance.

Cons: Expensive.

Amazon Kindle Scribe — Best e-reader E Ink tablet

Screen size: 11-inch | Battery life: Months | Network connectivity: Wi-Fi, Bluetooth | Capacity: Up to 64GB | Included stylus: Yes | Supported file types: AZW3, AZW, TXT, PDF, unprotected MOBI, PRC natively and more

The Kindle Scribe is the best choice if you want an E Ink tablet that doubles as an e-reader. The latest model has a larger 11-inch display, faster performance, and an improved front light system. It seamlessly integrates with the Kindle ecosystem, allowing you to mark up Kindle books (with some limitations) and take notes in dedicated notebooks. The included premium pen feels great and attaches magnetically. It's available in several configurations: a $430 base model without a front light, a $500 model with the improved lighting, and a $630 Colorsoft version with a color display. If you're invested in Amazon's ebook store, the Scribe is unmatched.

Pros: Fast, smooth performance; low-latency writing; integrates with Kindle books; premium pen included.

Cons: AI Search needs work; marking up Kindle books is inconsistent.

Supernote A6 X2 — Best E Ink tablet for writing and note-taking

Screen size: 7.8-inch | Battery life: Weeks | Network connectivity: Wi-Fi, Bluetooth | Capacity: 32GB, expandable up to 2TB with microSD card | Included stylus: No, extra cost | Supported file types: PDF, EPUB, Word, Text (.txt), PNG, JPG, Comic book archive (.cbz), FictionBook2 (.fb2) and OpenXPS (.xps)

The Supernote A6 X2 is a note-taker's dream. It has a 7.8-inch screen, a smooth FeelWrite 2 screen protector, and a software that prioritizes organization. It offers advanced features like keyword bookmarking, handwriting recognition, and a wide range of built-in templates. The device is also user-repairable with a replaceable battery and microSD expansion. The Manta A5 X2 is a larger 10.7-inch version if you need more screen space. The Nomad A6 X2 costs $329, but you'll need to buy a stylus separately (from $89). It's an excellent choice for bullet journaling and personal organization.

Pros: Excellent writing experience; tons of customization options; good handwriting recognition.

Cons: Pen costs extra; no backlight.

Onyx Boox Note Air4 C — Another color-display option

Screen size: 10.3-inch | Battery life: Weeks | Network connectivity: Wi-Fi, Bluetooth | Capacity: 64GB | Included stylus: Yes | Supported file types: PDF, PPT, EPUB, TXT, DJVU, HTML, RTF, FB2, DOC, MOBI, CHM, PNG, JPG, TIFF, BMP

The Onyx Boox Note Air4 C is a versatile Android-based E Ink tablet with a 10.3-inch color display. It runs Android 13, giving you access to the Google Play Store, so you can download apps like Kindle, Kobo, and even web browsers. The writing experience is low-latency, and the Notebook app includes useful features like colored brushes, shape insertion, and AI handwriting recognition. It supports many file formats and offers cloud sync via email, Google Drive, and more. At $529, it's on the expensive side, but its Android flexibility makes it a strong competitor.

Pros: Color display; runs Android 13 with Google Play Store access; supports many file-transfer methods; stylus included.

Cons: E Ink screen limits video use; not as user-friendly; expensive.

Onyx Boox Note Max — A big-screen option

Screen size: 13.3-inch | Battery life: Weeks | Network connectivity: Wi-Fi, Bluetooth | Capacity: 128GB | Included stylus: Yes | Supported file types: PDF, PPT, EPUB, TXT, DJVU, HTML, RTF, FB2, DOC, MOBI, CHM, PNG, JPG, TIFF, BMP

The Boox Note Max is perfect for those who want a large canvas for notes and reading. Its 13.3-inch E Ink Carta 1300 display is sharp and spacious, making it ideal for legal-pad lovers. It runs Android 13, so you have access to the Google Play Store, and it includes a stylus and cover in the box. The Calendar Memo app is particularly useful for logging daily notes and to-do lists. However, at $689, it's quite expensive and lacks a front light.

Pros: Sharp 13.3-inch display; Android 13; includes stylus and cover.

Cons: No front light; not user-friendly; expensive.

reMarkable Paper Pro Move — Honorable mention

Screen size: 7.3-inch | Battery life: Two weeks | Network connectivity: Wi-Fi | Capacity: 64GB | Included stylus: Yes | Supported file types: PDF, EPUB

The reMarkable Paper Pro Move is a compact version of the Paper Pro, with a 7.3-inch color display. It's ideal for those who want portability without sacrificing build quality or writing experience. It has the same software as the full-size Paper Pro, including support for PDFs and EPUBs, and the marker charges magnetically. At $449, it's cheaper than its bigger sibling but still pricey for its size.

Pros: Compact; excellent build quality; good battery life.

Cons: Expensive for a small tablet; no keyboard folio option.

Kobo Libra Colour — Honorable mention

Screen size: 7-inch | Battery life: Up to 40 days | Network connectivity: Wi-Fi, Bluetooth | Capacity: 32GB | Included stylus: No | Supported file types: EPUB, EPUB3, FlePub, PDF, MOBI, JPEG, GIF, PNG, BMP, TIFF, TXT, HTML, RTF, CBZ, CBR, Kobo audiobooks

The Kobo Libra Colour adds color to its 7-inch display and supports stylus input for note-taking and highlighting. It integrates with Dropbox and Google Drive, and its low-latency stylus makes it a surprisingly effective E Ink tablet. You can write in the margins of e-books, create notebooks, and convert handwriting to text. The stylus costs an extra $70, but the tablet itself is affordable at $219. It's a great option if you prefer Kobo's library and want a smaller, more portable device.

Pros: Color display; low-latency stylus; write in margins.

Cons: Smaller screen; stylus costs extra.

Are E Ink tablets worth it?

E Ink tablets are worth it for a specific group of people. If you love the feel of writing on paper but want digital organization, or if you want a distraction-free device for focused reading and note-taking, they're excellent. They are not ideal for video consumption or casual browsing, as E Ink screens have noticeable refreshes and limited color reproduction. However, for their intended purpose, they offer a unique and satisfying experience.

What to look for in an E Ink tablet

Writing and latency

The writing experience is the most important factor. Look for low latency and good pressure sensitivity. Many modern E Ink tablets offer near-instant response, but some are better than others. Check whether a stylus is included or if you need to purchase one separately.

Reading

Consider the screen size and supported file types. Larger screens are better for writing but may be uncomfortable to hold for long reading sessions. If you buy from Amazon or Kobo, your existing ebook library will be accessible. For third-party devices, check if they support EPUB, PDF, and other common formats.

Search functionality

Some E Ink tablets allow you to search handwritten notes, bookmark pages, and convert handwriting to text. This can be a huge productivity boost, especially for students or professionals who take extensive notes. Test the search features to see if they meet your needs.

Sharing and connectivity

Most E Ink tablets support Wi-Fi, and many offer cloud syncing or companion apps to access your notes elsewhere. Some support Dropbox, Google Drive, or email export. Consider how you want to use your notes after they're created and choose a device that offers the right connectivity options.

Price

E Ink tablets generally cost between $300 and $800. Prices have risen recently due to increased costs. The reMarkable Paper Pro went from $579 to $629, and Supernote and Onyx Boox have also raised prices. Decide on a budget and weigh the features you need most.

Other E Ink tablets we've tested

Onyx Boox Tab X C

The Boox Tab X C is a color version of the Tab X, with a 13.3-inch Kaleido 3 E Ink display, octa-core processor, and 6GB of RAM. It runs Android 13 and is one of the most powerful E Ink tablets available, but at $820, it's extremely expensive.

Lenovo Smart Paper

The Lenovo Smart Paper has good hardware but is tied to Lenovo's cloud service for full functionality. It offers solid Google Drive integration but is pricier than competitors and lacks software flexibility.

Onyx Boox Tab Ultra

The Boox Tab Ultra is an all-purpose Android tablet with an E Ink screen. It can browse the web, check email, and even watch YouTube, but the E Ink display makes these activities less enjoyable. It's best for those who prioritize reading and eye comfort above all else.

Source: Engadget News

Some handhelds are built for quick sessions and classic libraries, focusing on simplicity, long battery life, and pocketable designs. Others blur the line between console and PC, offering large screens, powerful chips, and access to massive game libraries at the cost of size, price, or endurance. There are also more experimental options that emphasize unusual controls or intentionally limited experiences.

After months of testing and tracking the fast-moving handheld space, these are the devices that stand out in an increasingly crowded field.

Best handheld gaming PC for most: Steam Deck

The Steam Deck remains the best balance of price, performance, and usability in the gaming handheld market. The Steam Deck OLED is a thorough upgrade over the original, featuring a 7.4-inch HDR OLED display that is brighter, faster, and more vivid than the entry-level model's 7-inch IPS panel. It supports HDR, has a 90Hz refresh rate, and offers up to 12 hours of battery life depending on the game. The OLED model starts at $549 for 512GB of storage.

The entry-level Steam Deck, at $399, delivers the same core experience with a more basic LCD display and a smaller 256GB SSD. It remains a strong budget pick. However, the Deck's hardware is aging in 2026. Many graphically demanding games released in the past couple of years don't run well, and Linux anti-cheat issues render some live-service games unplayable. Still, it can play countless titles that aren't possible on original handhelds at this price, from Elden Ring to Final Fantasy VII Rebirth. It's also superb for emulation.

Pros and cons

• Pros: Enough power for many modern PC games; user-friendly interface; vivid OLED display; excellent value on the LCD model; superb emulation performance.
• Cons: Bulky; not the most powerful hardware; doesn't officially support every Steam game or games from other PC clients.

Best premium handheld gaming PC: Lenovo Legion Go S (SteamOS, Z1 Extreme)

The Lenovo Legion Go S is the closest thing we have to a Steam Deck 2. It's the first third-party device to natively run SteamOS, offering the same conveniences as Valve's handheld but with more modern hardware. The configuration we tested includes an AMD Ryzen Z1 Extreme processor, 32GB of RAM, and a 1TB SSD. It can draw up to 33W in handheld mode or 40W plugged in, well above the Deck's 15W, making it better for resource-intensive games.

It has a larger 8-inch 1,920x1,200 display with a 120Hz refresh rate and variable refresh rate (VRR), but it's an LCD panel, not OLED, so colors aren't as vivid and there's no HDR. The Legion Go S is heavy and the fans are loud, but its ergonomics are good. The Z1 Extreme version costs $900. A lower-tier Z2 Go configuration is available for $650, but we haven't tested that yet.

Pros and cons

• Pros: Better performance than the Steam Deck; runs official SteamOS; spacious display with VRR; good ergonomics.
• Cons: Pricey; heavier than Steam Deck; no HDR support; weak haptics.

Best Windows gaming handheld: ASUS ROG Xbox Ally X

If you want more software flexibility, the ASUS ROG Xbox Ally X is our current favorite Windows-based handheld. It runs Windows 11, but ships with a new 'Xbox full screen experience' that uses a modified Xbox PC app as the default interface, making it easier to navigate with gamepad controls and aggregating games from multiple storefronts.

The Ally X is more powerful than the Steam Deck, thanks to a Ryzen Z2 Extreme chip and 24GB of RAM. Its 7-inch 1080p 120Hz display supports VRR, which keeps games smooth. It also has a huge 80Whr battery, lasting around three and a half hours in demanding games. Windows still has quirks on handhelds, but the flexibility to play games from any PC client is a major advantage. It costs $1,000.

Pros and cons

• Pros: More powerful than Steam Deck; works with any Windows gaming client; 1080p 120Hz display with VRR; comfortable grips; supports new full-screen Xbox experience.
• Cons: Expensive; a bit bulky; Windows 11 still has quirks; lackluster haptics; no included case.

Best mobile gaming handheld for most people: Retroid Pocket 5

The Retroid Pocket 5 is the handheld to get if you mainly want to emulate older consoles. It's an Android-based device powered by a Snapdragon 865 chip and 8GB of RAM, with a built-in fan for extra performance. It can play most PS2 and GameCube games at full speed, and most PSP, Dreamcast, PS1, and N64 games flawlessly. It can also handle Android games and cloud streaming.

It has a 5.5-inch 1080p OLED display, a sturdy design, Hall effect joysticks, and good speakers. The main downsides are the laborious setup required for emulators and a joystick layout that isn't ideal for modern games. At $219, it's excellent value.

Pros and cons

• Pros: Capable emulation and Android gaming performance; lovely OLED display; sticks and buttons feel great; sturdy, portable design.
• Cons: Requires a ton of tinkering to get some emulators working optimally; joystick layout isn't ideal for modern games.

Another good option: Retroid Pocket Flip 2

The Retroid Pocket Flip 2 is essentially the Pocket 5 in a clamshell shape. It runs on the same chip, has the same great OLED display, Hall effect joysticks, active cooling, and 5,000mAh battery. The clamshell design provides natural protection and makes it easier to throw in a bag. It costs $229.

However, the recessed joysticks can feel awkward, and the face buttons and d-pad are smaller than they should be. It's also thicker and heavier than the Pocket 5. If you prefer a DS-style clamshell, it's a solid buy.

Best mobile gaming handheld overall: AYN Odin 2

The AYN Odin 2 is a step up from the Retroid Pocket 5, thanks to its Snapdragon 8 Gen 2 processor. It can play everything the Retroid devices can, just smoother and more reliably. It handled all PS2 and GameCube games we tested at two to three times native resolution, and most 3DS and Wii games had little to no slowdown. It's also comfortable, with great battery life—more than eight hours for demanding emulation and over 20 hours for lighter tasks.

The 6-inch 1080p IPS display isn't as color-rich as an OLED, but it's bright and well-sized. It starts at $299, which is relatively pricey for a mobile handheld, but it's the best-performing option in its class.

Pros and cons

• Pros: Excellent emulation and Android gaming performance; comfortable; great battery life.
• Cons: Relatively pricey; setting up emulators can still be laborious; docked experience isn't seamless.

Best Game Boy-style handheld: Analogue Pocket

The Analogue Pocket is the ultimate Game Boy. It plays authentic Game Boy, Game Boy Color, and Game Boy Advance cartridges via FPGA emulation, which is near-perfect with excellent responsiveness. It has a gorgeous 3.5-inch display with multiple filter modes, a sturdy build, and support for optional adapters for other retro systems. It can also run ROMs from a microSD card. At $220, it's not cheap, and stock issues are common, but it's the most elegant way to play classic cartridges.

Pros and cons

• Pros: Near-perfect cartridge emulation; gorgeous display; impressive build quality; expandable via adapters; supports software emulation.
• Cons: Stock issues and shipping delays are common; shoulder buttons feel spongy; tiny volume buttons.

What about the Nintendo Switch 2?

The Nintendo Switch 2 is more popular than any of the handhelds above, but it exists in its own world. The main reason to buy a Switch 2 is to play new Nintendo games, which no other device can legally offer. It has a significantly improved 7.9-inch display, magnetic Joy-Con controllers, and more storage. However, the list of true must-play exclusives is still limited a few months after launch. Unless you're worried about tariff-induced price hikes, you don't need to rush out and buy one just yet.

Other gaming handhelds we've tested

The Lenovo Legion Go 2 is a capable alternative to the ROG Xbox Ally X with a vivid 8.8-inch OLED display and detachable controllers, but it's heavy and expensive, starting at $1,100. The MSI Claw 8 AI+ offers strong performance and battery life, but at $1,100, it's hard to justify over the Ally X. The ModRetro Chromatic is a premium Game Boy-style handheld, but it requires AA batteries and lacks save states, making it a tough sell at $200.

Other notable devices include the Ayaneo Flip DS, a dual-screen Windows handheld that's more of a neat idea than a fully realized product, and the Ayaneo Kun, a powerful but expensive Windows handheld. On the budget side, the Retroid Pocket Classic offers great value at $129 with a superb OLED display, while the Miyoo Mini Plus and TrimUI Brick are even cheaper options with compromises. The PlayStation Portal is a niche accessory for PS5 streaming, and the Logitech G Cloud, while comfortable, is overpriced compared to the Retroid Pocket 5.

What to know about the gaming handheld market

The market can be broken into three tiers. At the top are x86-based portable gaming PCs like the Steam Deck and ROG Ally X, which are the most powerful but also the largest, most expensive, and have shorter battery life. The middle tier includes Android- or Linux-based mobile handhelds like the Retroid Pocket, which are more compact and affordable but can't run modern PC or console games, making them ideal for emulation and cloud streaming. The final tier includes devices like the Switch 2 and Playdate that run custom software and offer unique game libraries.

Recent updates to this guide include the addition of the ASUS ROG Xbox Ally X as our top Windows pick, testing notes on several new emulation handhelds, and the addition of the SteamOS version of the Lenovo Legion Go S. We're also keeping an eye on newer releases like the Retroid Pocket 6, AYN Odin 3, and other upcoming devices.

Source: Engadget News

In an era when consumers are bombarded with endless choices, tech reviews serve as a crucial guide. A thoughtful, transparent and honest review can help people decide whether a product is worth their hard-earned money. That is why we take our reviews policy seriously. Whether the product is a smartphone, a laptop, a robot vacuum or a smart home device, our goal is to provide a fair, balanced and informative critique that gives readers an accurate representation of what they can expect from each product.

With no shortage of ways to spend your money on new technology, we want to help readers make smart, informed decisions about the value of each product at any given price point. We cover a wide range of technology products, with preference given to unique features and emerging technologies. We always review the latest models from major brands, but we also strive to unearth and elevate niche products worthy of consumer attention. Each product is judged based on real-world experience, and, except where clearly noted, all of our product reviews are based on market-ready products, not pre-release prototypes.

Given the speed at which new products are released and existing products are updated, it is important to remember that a review is a snapshot and that the competitive landscape can change over time. Today's best-in-class device might be outclassed within a few months. Still, we provide honest and thorough assessments of every product we review, considering both the good and bad aspects. While the criteria for what makes a successful product will vary, our goal is always to indicate its overall quality and value.

Review Philosophy

Our reviews typically include a mix of quantitative and qualitative testing. We run different benchmarks depending on the type of product, but we always approach reviews as consumers because we are consumers. We are early adopters, tinkerers and generally tech-obsessed. And while we definitely spend more time than most with things like smartphones, laptops and robot vacuums, we still have questions when a new device comes across our desks: Does it deliver on all of its promises? Is it better than the next best thing out right now? Who should buy it? Or more importantly, is it worth buying at all?

In order to answer those questions, we aim to use the products we review as if they were our own. For example, we might carry a smartphone everywhere for days, taking selfies, playing games and annoying coworkers; or we will write a laptop review on that machine to see how the keyboard holds up over extended typing sessions. For things like synths and samplers, we use them for as long as we can so we can judge how usable the device is, both for seasoned creators and novices. Throughout the process, we make notes on build quality, versatility and overall value.

We know that gadgets do not exist in a vacuum, so placing them into proper context is just as important as judging their performance and features. That involves not only comparing a device to its direct competitors, but also thinking critically about who would get the most use out of it and whether the value it provides justifies its price. No gadget is perfect, but one might be perfect for your specific needs. That is why we aim to be as detailed and practical as possible, so readers can match a product to their own priorities.

Different reviewers may have very different opinions about a particular product, depending on personal preferences, experiences or tastes. It is important to remember that a review reflects just one person's opinion, and one moment in time. We encourage readers to look at multiple reviews and gather as much information as possible before making a purchase. Our job is not to make the decision for you, but to arm you with the insights you need to make the best choice.

Scoring

We rate products on a 1-100 scale, and the final score is chosen by our editorial team with no outside input. Each product's score is agreed upon by the reviewer and their editor, sometimes with advice from team members who have expertise in the area. The score is not influenced by advertisers, manufacturers or any external party. It is a direct reflection of how well the product performed during testing and how it compares to competing devices in its category.

What our scores mean

• 0-29: An awful product that should be avoided. You will be sorry.
• 30-49: A poor product that is more likely to be frustrating than useful.
• 50-54: A disappointing product with a lot of problems.
• 55-59: It might work in a pinch, but you would not want to use it regularly.
• 60-64: A forgettable product. It is not bad, exactly, but also does not do anything particularly well.
• 65-69: It has some redeeming qualities, but you could do better.
• 70-74: A dependable product that does not stand out from the competition or is very niche in appeal.
• 75-79: A very good product that falls just short of greatness, but we can still heartily endorse.

Recommended (80-89)

• 80-84: There is a lot to like about this in spite of a few flaws. Easy to recommend to most shoppers.
• 85-89: An all-around great product that is among the best in its category. You will almost certainly be happy.

Editors' Choice (90-100)

• 90-94: The best in its category. We highly recommend it.
• 95-99: An important, nearly flawless product that raises the bar for the rest of the tech industry.
• 100: A gadget unicorn.

Unscored Reviews

Our review scoring system is designed for the products core to our coverage. From time to time, you will see reviews with no score. We do not numerically rate works of art like movies or games; our media reviews instead comprise a mix of opinion, criticism and analysis. When we review software, cars and other products outside of the typical consumer electronics market, we do not apply a score. This allows us to adapt our evaluation method to the unique qualities of each product and provide a more nuanced critique. A lack of score does not mean a lack of rigorous testing; it simply means that a numerical rating would not be appropriate.

Embargoes and Review Units

We choose which products to review independent of our parent company or advertisers. Review units are usually provided by companies free of charge, which is standard practice in the industry. In most cases, at the end of our review period, hardware is returned to the company. Occasionally we may hold onto units for long-term testing, but under no circumstances do they become a reviewer's personal property. Reselling of review units is strictly prohibited. This policy ensures that there is no financial incentive to inflate a review or to give a product more favorable coverage in exchange for a free sample.

We believe in transparency. While our reviews are typically based on products supplied by manufacturers, as is the standard practice in the industry, our opinions and conclusions are independent. Where applicable, we accept early products so that we can give as thorough and informative a review as possible. This process allows us to test a device thoroughly before the general public gets their hands on it, which helps consumers make informed pre-order or early purchase decisions. It also lets manufacturers know what issues may exist before a product hits the market, though we do not share our results with them prior to publication.

We are committed to maintaining the trust of our readers. That means avoiding conflicts of interest, being open about how products are sourced, and making our scoring methodology clear. Every review we publish is the result of a collaborative editorial process that prioritizes facts, context and real-world usability. We never submit to pressure from companies to change a review or a score, and we always disclose if there are any limitations to our testing. This policy is part of our broader commitment to journalistic integrity and to serving the best interests of our audience.

Source: Engadget News

L'événement a eu lieu au sein du palais apostolique de Castel Gandolfo, la résidence d'été traditionnelle des papes, située à une trentaine de kilomètres de Rome. Le pape Léon XIV a présidé cette célébration particulière, voulue comme une invitation à la paix dans le monde. La présence d'Andrea Bocelli, artiste connu bien au-delà des frontières italiennes, a donné à cette cérémonie une dimension émotionnelle forte, mêlant foi, musique et universalité. Les 164 choristes qui l'accompagnaient ont été spécialement réunis pour l'occasion, créant un ensemble vocal impressionnant, capable de porter un message de fraternité à la hauteur de l'enseignement de saint François.

Un message de paix porté par la musique

La prière pour la paix a été pensée comme un temps de silence, de méditation et de chant. Dans son homélie, le pape Léon XIV aurait rappelé que la paix ne se construit pas seulement entre les nations, mais aussi dans le quotidien des hommes, dans leurs gestes, leurs paroles et leurs relations. Andrea Bocelli, de son côté, a interprété plusieurs pièces, dont certaines sont devenues des classiques du répertoire sacré et lyrique. Sa voix puissante, à la fois limpide et vibrante, a porté les mots de l'espérance au milieu des jardins de la résidence pontificale, devant un public restreint composé de religieux, d'autorités locales et de fidèles venus spécialement d'Assise.

Le choix d'Assise n'est pas anodin. C'est dans cette ville ombrienne que saint François est né à la fin du XIIe siècle, et c'est là qu'il a choisi de vivre une vie de pauvreté, de prière et de service aux plus démunis. Il a fondé l'ordre des Frères mineurs, plus connu sous le nom de franciscains, et son rayonnement spirituel a traversé les siècles. Mort dans la nuit du 3 au 4 octobre 1226, il a été canonisé dès 1228 par le pape Grégoire IX. Huit siècles plus tard, sa figure continue d'inspirer non seulement les catholiques, mais aussi les croyants d'autres religions et les hommes de bonne volonté.

Saint François, un saint universel

François d'Assise est souvent présenté comme le saint le plus populaire d'Italie, et l'un des plus aimés au monde. Il est le saint patron des animaux, de l'écologie et de l'Italie, mais il est aussi un modèle de dialogue interreligieux. En 1219, il aurait rencontré le sultan Malik al-Kamil pendant la cinquième croisade, un épisode devenu symbole de rencontre pacifique avec l'islam. Sa prière, connue sous le nom de « Cantique des créatures », est une louange à Dieu pour la beauté du monde naturel, souvent citée comme une anticipation de la sensibilité écologique moderne. C'est d'ailleurs en référence à cette dimension que le pape François, premier pape de l'histoire à choisir ce nom, a publié l'encyclique Laudato si' sur la sauvegarde de la maison commune.

La célébration du huitième centenaire de sa mort a donc revêtu une importance particulière. Tout au long de l'année 2026, des événements religieux et culturels se sont succédé en Italie et dans le monde pour honorer la mémoire du Poverello, comme on le surnomme affectueusement. Des pèlerinages, des conférences, des expositions et des concerts ont été organisés, avec un fil conducteur : rappeler l'actualité du message franciscain face aux défis contemporains, qu'il s'agisse de la paix, de la justice sociale ou de la protection de l'environnement.

Andrea Bocelli, une voix pour l'éternité

Andrea Bocelli n'en est pas à son premier geste de ce type. Depuis des décennies, il met sa notoriété au service de causes humanitaires et de moments de recueillement. Chanteur aimé d'un large public, il a vendu plus de 80 millions de disques à travers le monde et s'est produit dans les plus grandes salles et les plus prestigieux festivals. Son album Sacred Arias, paru en 1999, s'est vendu à plus de 5 millions d'exemplaires et est devenu l'un des albums de musique classique les plus populaires de tous les temps. Ses collaborations avec des artistes comme Céline Dion, Sarah Brightman ou Ed Sheeran ont contribué à faire de lui une figure internationale, capable de franchir les frontières entre la musique classique et la chanson populaire.

La présence de Bocelli dans un cadre aussi solennel que la résidence pontificale n'est pas seulement un événement médiatique. Elle témoigne de la place que la musique occupe dans la liturgie catholique et dans la vie de l'Église. Le chant peut être une prière, une manière d'élever le cœur et de rassembler les personnes dans une même émotion. Les 164 choristes présents aux côtés du ténor illustrent cette idée : aucun soliste ne porte seul le message, il faut une communauté, une unité des voix, pour que l'harmonie devienne une expérience spirituelle.

Un concert dans les jardins pontificaux

Les jardins de la résidence d'été des papes, avec leurs allées ombragées, leurs fontaines et les vues sur le lac d'Albano, offraient un cadre idéal pour cette méditation musicale. Les images diffusées par les médias du Vatican ont montré Andrea Bocelli debout, les yeux fermés, tandis que la chorale s'élevait autour de lui. Le répertoire comprenait des chants traditionnels italiens, des pièces sacrées et probablement l'Ave Maria de Schubert, qui fait partie des airs que le ténor a le plus souvent interprétés dans les cérémonies religieuses.

La retransmission de l'événement a permis à des millions de personnes de suivre ce moment de prière à distance. Sur les réseaux sociaux, de nombreux internautes ont partagé leur émotion, saluant la beauté de la musique et la profondeur du message. En ces temps de tensions politiques et militaires, ce type de manifestation peut sembler symbolique, mais il rappelle que le dialogue et la paix passent aussi par la culture et l'art. La musique a cette capacité unique de toucher le cœur sans passer par le filtre des idéologies ou des intérêts.

Un centenaire chargé d'histoire

Le huitième centenaire de la mort de saint François d'Assise a été officiellement ouvert le 3 octobre 2025, avec l'année de préparation qui a conduit aux célébrations principales de 2026. De nombreux événements ont eu lieu dans la ville d'Assise, notamment la basilique Saint-François, classée au patrimoine mondial de l'UNESCO, qui abrite les reliques du saint et les fresques de Giotto. L'événement du 29 juillet a toutefois eu lieu dans un cadre plus privé, la résidence d'été des papes, ce qui lui a donné un caractère plus intime et solennel à la fois.

Le choix de la date n'a pas été laissé au hasard. Le 29 juillet est proche de la fête de sainte Marthe, mais surtout il permettait de réunir les pèlerins et les représentants de différentes confessions avant les grandes rencontres prévues à l'automne, autour du 3 octobre, date anniversaire de la mort du saint. En présidant cette prière, le pape Léon XIV a souhaité montrer que la paix reste une priorité du pontificat. Dès les premières semaines de son mandat, il avait exprimé son intention de faire du dialogue et de la réconciliation les axes majeurs de son enseignement.

La musique sacrée comme pont entre les cultures

La musique sacrée a toujours occupé une place particulière dans l'histoire de l'Église. Depuis le chant grégorien jusqu'aux grandes œuvres de Bach, de Mozart ou de Verdi, la tradition musicale chrétienne est immense. Andrea Bocelli, avec sa carrière internationale, s'inscrit dans cette longue lignée interprète capable de rendre accessibles des œuvres que certains jugent élitistes. Son concert devant le pape Léon XIV a montré que la musique sacrée n'est pas un genre muséal, mais une force vivante, capable de parler aux hommes et aux femmes du monde contemporain.

Les 164 choristes, venus pour certains de très loin, ont passé plusieurs jours de répétition avant cette représentation. Leur travail collectif, dirigé par un chef de chœur, a été salué par le public et par les organisateurs. La direction artistique a veillé à ce que chaque détail soit parfait : les placements vocaux, l'équilibre des voix, les nuances sonores. Le résultat fut un moment d'une grande intensité, retransmis en direct par des chaînes de télévision italiennes et internationales.

Un message d'espérance après un mois tumultueux

Cette cérémonie intervient après plusieurs semaines marquées par des crises internationales, des conflits persistants et des inquiétudes sur l'avenir du climat. La prière présidée par le pape Léon XIV a pris une dimension prophétique en ce sens. En invitant un artiste aussi populaire qu'Andrea Bocelli, l'Église catholique cherche à montrer qu'elle ne se replie pas sur elle-même, mais qu'elle souhaite s'adresser au plus grand nombre. La musique, plus que le discours, permet cette rencontre directe avec l'émotion et le cœur.

Andrea Bocelli, qui a souvent dit que sans la foi il n'aurait pas surmonté les épreuves de sa vie, a lui-même témoigné de l'importance de ces moments dans sa carrière. Il est né avec un glaucome congénital et est devenu complètement aveugle à l'âge de 12 ans, après un accident de football. Son parcours exceptionnel, de l'enfance modeste en Toscane aux plus grandes scènes du monde, est une source d'inspiration pour beaucoup. Dans cette perspective, sa présence à Castel Gandolfo n'était pas seulement la performance d'un grand artiste, mais aussi un témoignage de résilience et d'espérance.

L'événement s'est achevé par une bénédiction du pape et l'interprétation d'un chant final, repris par l'ensemble des participants. Selon les témoins, un silence émouvant a suivi la dernière note, avant les applaudissements. Le pape Léon XIV est ensuite allé saluer le ténor et les principaux choristes, les remerciant pour ce moment de grâce. La soirée s'est poursuivie de manière plus informelle, tandis que la nuit tombait sur les collines des Castelli Romani.

Pour les fidèles, cette journée restera gravée comme un exemple de ce que l'Église peut offrir au monde : un espace de paix, de beauté et de recueillement. Pour les amateurs de musique, le concert restera comme une nouvelle illustration du talent d'Andrea Bocelli. Pour les observateurs politiques, il aura au moins montré que les symboles comptent encore dans les relations internationales. Peut-être est-ce là, justement, le sens profond de l'héritage de saint François : rappeler que la fraternité n'est pas une utopie, mais une tâche quotidienne, à laquelle chacun peut contribuer, même par une chanson.

Source: Le Figaro News

« Je suis vraiment désolé, pour tous ceux qui m'ont témoigné de l'amour et ont cru en moi, pour l'inquiétude et la déception que j'ai causées », a déclaré l'acteur de 51 ans, visiblement ému face à l'assemblée. « Je vais accepter et endurer toutes les critiques. En tant que père, je vais assumer mes responsabilités envers mon fils jusqu'à la fin », a-t-il renchéri, avant de quitter la scène sous les applaudissements polis mais tièdes du public.

La veille, son agence Artist Company avait confirmé qu'il était bien le père du fils de la mannequin Moon Ga-bi, né en mars dernier. La jeune femme de 35 ans avait annoncé récemment être devenue mère, sans révéler l'identité du père, mais les spéculations allaient bon train sur les réseaux sociaux. Selon un média local, Moon Ga-bi aurait souhaité se marier avec Jung Woo-sung afin de « donner une famille à son enfant », mais l'acteur aurait refusé cette demande.

Ce refus présumé a immédiatement été qualifié d'« irresponsable » par de nombreux internautes et commentateurs, d'autant que la Corée du Sud reste un pays socialement conservateur où les mères seules et leurs enfants sont encore fortement stigmatisés. Bien que Jung Woo-sung ait assuré qu'il agirait comme un père, son silence autour de la question du mariage a laissé un goût amer. Dans un pays où le mariage est souvent considéré comme le seul cadre légitime pour élever un enfant, la décision de l'acteur est perçue par beaucoup comme un refus de s'engager pleinement.

Une carrière longtemps irréprochable

Né en 1973, Jung Woo-sung est l'un des acteurs les plus respectés du cinéma sud-coréen. Révélé dans les années 1990, il a enchaîné les succès critiques et commerciaux, de « The Good, the Bad, the Weird » à « Steel Rain », en passant par « Innocent Witness », qui lui a valu de nombreux prix. Il a également travaillé avec des réalisateurs de renom et a été salué pour son engagement humanitaire. Il a notamment été ambassadeur de bonne volonté de l'Agence des Nations unies pour les réfugiés (UNHCR) pendant près de dix ans, jusqu'en juillet dernier, parcourant le monde pour attirer l'attention sur le sort des déplacés et des réfugiés, et visitant des camps en Somalie, au Yémen, au Bangladesh ou encore au Liban.

Cette casquette d'humanitaire lui a souvent valu des éloges, mais elle s'est retournée contre lui après le scandale. Sur les réseaux sociaux et les forums, des détracteurs ont vivement critiqué son comportement, estimant que sa posture morale était en décalage avec sa vie privée. « Il a tellement parlé de l'accueil des réfugiés et pourtant il a fait de son propre fils un réfugié », a commenté un utilisateur sur le site d'information local. Cette pique, reprise massivement, illustre la sévérité du jugement moral qui s'est abattu sur l'acteur.

Des enfants souvent stigmatisés

Cette affaire intervient dans un contexte où la Corée du Sud connaît une crise démographique sans précédent. Le pays affiche le taux de natalité le plus bas du monde, avec seulement 0,72 enfant par femme en 2023. Les naissances hors mariage y représentent à peine 4,7 % du total, soit l'un des chiffres les plus bas parmi les 38 pays de l'OCDE, où la moyenne se situe autour de 40 %. En comparaison, en France, environ 60 % des naissances se font hors mariage, et même dans des pays asiatiques voisins comme le Japon, ce chiffre dépasse les 2 % mais reste bien inférieur aux standards occidentaux. En Corée du Sud, le faire et le dire restent profondément tabous.

Ce tabou s'explique en partie par une définition légale très stricte de ce qu'est une famille. Selon les experts, le droit sud-coréen ne reconnaît quasiment que la famille fondée sur le mariage hétérosexuel. Les adoptions par un adulte non marié sont extrêmement rares, les femmes célibataires ne peuvent pas recourir à un don de sperme, et le mariage homosexuel n'est pas reconnu. Ces obstacles juridiques, hérités d'une vision patriarcale et conservatrice de la société, dissuadent de nombreuses personnes de fonder une famille en dehors du cadre conjugal. Ils expliquent aussi pourquoi les mères célibataires sont souvent poussées à l'isolement ou à l'abandon de leur enfant.

Le gouvernement sud-coréen, conscient de l'urgence démographique, tente de multiplier les incitations au mariage et à la natalité. Séoul a récemment décidé de mettre à disposition des futurs mariés une vingtaine de lieux de cérémonie gratuitement, espérant ainsi faciliter les unions. Mais ces mesures peinent à enrayer la chute de la nuptialité et de la natalité, tant les causes sont profondes et structurelles.

Dans ce contexte, la sortie de Moon Ga-bi et le scandale Jung Woo-sung ont rouvert un débat nécessaire sur la diversité des modèles familiaux. La parlementaire d'opposition Lee So-young, membre du Parti démocrate, a pris la parole pour défendre une vision plus ouverte de la famille. Ayant elle-même vu ses parents divorcer lorsqu'elle était enfant, elle a insisté sur le fait que « la réalité est que chacun est unique » et qu'une « société qui respecte ces différences serait sans aucun doute une meilleure société, non ? ». Une intervention qui tranche avec le discours dominant, mais qui reflète l'évolution des mentalités, surtout chez les jeunes générations.

En attendant, Jung Woo-sung devra faire face aux conséquences de ses actes. Son agence a annoncé qu'il assumerait pleinement son rôle de père, et l'acteur semble déterminé à ne pas esquiver. Mais l'opinion publique reste divisée. Certains saluent son courage de reconnaître l'enfant et de s'excuser publiquement, d'autres estiment que ses excuses arrivent trop tard et que son refus du mariage en dit long sur sa conception de la famille. L'affaire n'est pas près de s'éteindre, et elle pourrait avoir des répercussions sur sa carrière, sa réputation et son image de héros romantique.

Source: leparisien.fr News