H.Headline
Chaos

THE EDITION ARCHIVE

A front page, saved in time.

Making AI Headlines · AM edition

Published · New York time

Permanent edition link →

10 pages · 23 stories

PAGE 1 OF 10

Claude Rejects Game Clones, Leaving Knockoffs to Find Another Coworker

Anthropic restricts recreations of existing games, while publishers take a privacy browser to court.

Making AI Headlines newspaper, page 1: Claude Rejects Game Clones, Leaving Knockoffs to Find Another CoworkerPage 1 of 10
Page 1 of 10. Read the same stories below.

Claude Gets a Rulebook for Games It Cannot Recreate

Anthropic is blocking Claude Code from recreating existing video games after a wave of AI-built knockoffs. Users can still ask Claude to help design original games or games in an existing genre, according to the reporting. The company has drawn a line between borrowing a broad idea and asking an assistant to rebuild an existing title. That line gives game makers a little more room to breathe and gives prompt writers a new puzzle. The assistant is meant to help people make software quickly, but speed is less appealing to the creator of a game when the requested destination is a copy. Anthropic's answer is to put a limit inside the product rather than leave every dispute for a court to settle. The rule also has a reach problem. A restriction on one coding assistant does not remove other tools or erase the appeal of a quick imitation. Original developers must still compete in a market where software can make copying easier, while AI companies must decide how much friction to add to products sold on reducing it. For Claude, the new assignment is to help build a game without becoming the world's most efficient arcade counterfeiter. The creative brief now comes with a small but consequential addition: please make something of your own.

Share articleOpen article permalink →

Publishers Tell Brave Its Privacy Pitch Does Not Cover Their Articles

The New York Times and other publications have sued Brave, alleging that the privacy-focused browser company scraped copyrighted news content for AI uses. The Times alleges that Brave unlawfully distributed its material to customers including Anthropic and Perplexity. Other publishers also allege that Brave enabled users to bypass their paywalls. The claims put a browser known for protecting users' privacy at the center of a dispute over who may collect and distribute publishers' work. They are allegations in lawsuits, not findings that Brave broke the law. The publishers want control over material that costs money to report and edit, while AI services have an appetite for information they can deliver in new forms. The dispute goes beyond a single browser window. If a service can provide material without sending a user to the publisher, the value of the original visit becomes part of the argument. Paywalls, licensing and AI-generated answers all sit uncomfortably close together. Brave's marketing asks people to consider who can see their browsing. The publishers are asking a different question: who gets to use the articles once someone has found them? A privacy setting is not much help with that bill, and the next stop is a courtroom rather than a preferences menu.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 2 OF 10

OpenAI Tidies Its Tabs and Rebrands the Building Behind Them

A desktop redesign promises one workspace, while a possible data center makeover starts with the name.

Making AI Headlines newspaper, page 2: OpenAI Tidies Its Tabs and Rebrands the Building Behind ThemPage 1 of 10
Page 2 of 10. Read the same stories below.

Chat and Work Unite, Bringing Their Usage Meter to the Meeting

OpenAI plans to merge the Chat and Work tabs in its ChatGPT desktop app, first offering the new experience to select users for feedback before a wider rollout. The company says ChatGPT and Codex will draw usage from the same allocation pool after the change. The current app has separate ChatGPT and Codex sections, and the ChatGPT section has its own Chat and Work tabs. That arrangement can make it difficult to remember where a project lives. OpenAI says the combined experience will let users move from exploring an idea to having ChatGPT carry it through in one place, with their files brought along. The change is a practical attempt to make a complicated product easier to navigate. It also puts two activities under a shared usage allocation. People who move between conversation and coding will have one less interface boundary to cross, but they will still need to keep track of how much of their allowance remains. A single workspace might spare a user the daily expedition through overlapping tabs. It cannot make every project simple, but it can at least stop the interface from adding a small navigation test before the work begins. The tabs are being consolidated; the bill for activity is coming to the same meeting.

Share articleOpen article permalink →

A Data Center Applies to Become a Campus Without Moving

OpenAI may call its data centers AI compute campuses as it prepares a public relations push around the political debate over artificial intelligence. TechCrunch, citing Axios, reports that chief global affairs officer Chris Lehane expects AI to be a major issue in the 2028 presidential race and has discussed proposals including AI speed limits. The possible name change arrives as data centers face public concern over their demands on local resources. A campus sounds considerably more collegial than a warehouse full of computing equipment. The machines, however, do not draw less power because the sign outside has acquired an academic air. OpenAI's broader effort aims to shape how voters and officials think about the technology, including whether people see AI as empowering them. That is a larger task than choosing a label. Communities considering new facilities have reasons to ask about electricity, water and the costs and benefits of development, whatever a company calls the site. A rebrand can change the first sentence of a presentation. It cannot answer the questions that come after it. The proposed campus may yet get its polished brochure, but the neighbors are likely to bring their own reading list, starting with the power bill and the local permitting documents.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 3 OF 10

AI Infrastructure Discovers Courts and Computers Both Have Rules

A Memphis pollution case moves forward, and OpenAI increases rewards for security reports.

Making AI Headlines newspaper, page 3: AI Infrastructure Discovers Courts and Computers Both Have RulesPage 1 of 10
Page 3 of 10. Read the same stories below.

Memphis Turbines Meet a Permit Question That Will Not Dismiss Itself

A federal judge denied xAI's request to dismiss an NAACP lawsuit alleging that its Memphis data center campus runs turbines without proper permits in violation of the Clean Air Act. US District Judge Debra M. Brown's decision allows the case to proceed; it does not decide whether the allegations are true. The NAACP filed the action in April, accusing the company, now known as SpaceXSI, of using unpermitted methane gas turbines to power data centers in the Memphis area. The dispute places the infrastructure behind AI models in a familiar local setting: a community, an industrial power source and a disagreement over the rules governing it. The company wants computing capacity to run its systems. The organization bringing the case wants the court to examine whether the power supply complies with clean air requirements. Those demands operate on different clocks, and a desire for rapid computing growth does not make a permitting question disappear. For a business devoted to quick answers, the court has delivered a slower process. The next stage will turn on evidence and legal arguments, not on how fast a model can produce text. A data center can make its equipment work around the clock, but the lawsuit still has to work its way through court.

Share articleOpen article permalink →

OpenAI Makes Finding Its Bugs a Better-Paid Outside Job

OpenAI is quadrupling bug bounty payouts, with researchers able to earn up to $172,000, and launching a new internal security sprint. The program gives independent researchers a financial reason to look for weaknesses and report them to the company. A bounty is a practical bargain: outsiders test products from different angles, and the company gets a chance to address qualifying problems. The top figure is a maximum, not a promise that every report earns six figures. OpenAI is also working on its defenses internally, so the outside program complements rather than replaces its own security work. The larger rewards make an attractive advertisement for researchers whose skill is finding the things a product team would rather have caught first. In ordinary office life, pointing out a colleague's mistake may earn a tense meeting. In a bug bounty program, the right discovery can come with a check. The arrangement still depends on more than the size of the advertised prize. Useful reports need to reach the company, and identified problems need attention. The security sprint and the bounty both suggest that OpenAI expects protecting its systems to require continuous work. The digital house has an internal repair crew, but it is also paying visitors to inspect the locks.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 4 OF 10

AI Agents Get a Front Desk While Watches Take Meeting Notes

Uber adds controls for thousands of tools, and Apple tests conversation features for the wrist.

Making AI Headlines newspaper, page 4: AI Agents Get a Front Desk While Watches Take Meeting NotesPage 1 of 10
Page 4 of 10. Read the same stories below.

Uber Gives Its Agents a Bureaucracy Before They Touch the Tools

Uber has built an MCP Gateway to standardize how AI agents discover and invoke internal services across more than 800 MCP servers and 5,000 tools. The gateway separates tool registration from runtime execution and includes automated API discovery, access controls and protections for sensitive data. The scale explains the need for a checkpoint. An agent that can call an internal service needs to know what is available, but the company also needs limits on which services it can use and what information passes through. The gateway includes sensitive-data redaction and response filtering, along with incremental tool discovery through Omni MCP. For developers, the design offers a more consistent way to manage a large collection of services. For the business, it puts permissions between an automated request and the systems that might carry it out. It is less glamorous than an agent completing a task in seconds, but it addresses the awkward question of what the agent should be allowed to do first. The comic achievement is a front desk for software hired to eliminate administrative chores. The agents can discover thousands of tools, provided they pass through the proper gateway. Automation has arrived at the office and immediately been handed a permissions form.

Share articleOpen article permalink →

Apple Watch Offers to Remember the Conversation Humans Delegated

Apple's Audio Intelligence features for Apple Watch have entered beta testing, with tools intended to help users remember what someone said or take notes on conversations. The feature puts an AI-assisted recap on a device already attached to the person trying to keep up. The appeal is straightforward. Meetings and conversations can move quickly, and a useful note or reminder may save someone from losing an important detail. On a watch, the promise is especially compact: a small device becomes a place to retrieve information that did not stay in a person's memory. The test also gives Apple a way to explore how much assistance people want from a device during ordinary conversation. A recap is not the same thing as participating in the exchange, and users will still have to judge what matters. The beta label means the features are being tested rather than presented as a finished answer to every missed detail. The joke is that technology designed to rescue attention might become another reason to outsource it. A person can nod earnestly while trusting the watch to keep the minutes. At the end of the meeting, the best-informed participant may be the one with no opinion at all, strapped quietly to someone's wrist.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 5 OF 10

AI Predicts Your Next Move, While Apple Leaves Its Revenue Move Unspecified

A startup exits stealth with a behavior-modeling claim, and Apple hints at third-party AI opportunities.

Making AI Headlines newspaper, page 5: AI Predicts Your Next Move, While Apple Leaves Its Revenue Move UnspecifiedPage 1 of 10
Page 5 of 10. Read the same stories below.

42AI Raises $62.5 Million to Forecast Decisions Not Yet Made

42AI has emerged from stealth with $62.5 million in financing and a world model that the startup says can predict human behavior, even in new situations. The claim puts a price tag on an ambitious idea: using AI to anticipate choices in settings the system has not encountered before. Predicting behavior is an appealing business proposition because organizations routinely make plans under uncertainty. A model that helps narrow that uncertainty could attract customers, but the startup's claim is not the same thing as proof that every person's decisions can be forecast. The useful questions are how well the system performs and under what conditions. The company's funding announcement puts it among startups seeking to turn AI models into tools for interpreting the world, rather than simply producing text. The novelty is the promise that the model can extend its predictions into unfamiliar situations, where familiar patterns may be less reliable. There is also a fine irony in selling certainty about human choice. People already spend plenty of time trying to guess what other people will do, often with mixed results and strong confidence. Now a well-funded model wants a seat at that table. Its first customer may be tempted to ask whether buying the model was predictable too.

Share articleOpen article permalink →

Apple Finds a Third-Party AI Opportunity and Keeps the Details

Apple hinted during its Monday night earnings call that third-party AI app opportunities could help spur revenue. The remarks offered a glimpse of the company's interest in earning money from AI beyond its own products, without laying out a detailed new business model. Apple controls a large platform on which outside developers distribute apps and services. That makes third-party activity a natural subject for investors assessing how AI might affect its business. But an opportunity is not yet a disclosed product, fee or revenue forecast, and the company's hint leaves the mechanics open. For developers, the question is how any future arrangement would work and what share of the value would reach the people building the apps. For Apple, the attraction is a platform where other companies create services that bring customers and spending into its ecosystem. Both sides can see the potential, even if the earnings call did not supply a complete blueprint. The comic center is a familiar platform-company talent: turning a broad possibility into a topic for close financial attention. Apple has not had to unveil a device, announce a price or put anything in a box. It has merely suggested there may be an opportunity, and the opportunity has already acquired a very interested audience.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 6 OF 10

The AI Workday Gets Longer, and the Antibiotic Search Gets a Prize

Garry Tan puts a price on heavy agent use as two MIT researchers receive recognition.

Making AI Headlines newspaper, page 6: The AI Workday Gets Longer, and the Antibiotic Search Gets a PrizePage 1 of 10
Page 6 of 10. Read the same stories below.

Garry Tan's Agents Work Late, and Their Token Bill Keeps Pace

Y Combinator's Garry Tan says he spends up to $200,000 a year on AI tokens as he works with agents. Speaking on Walter Isaacson's podcast The Builders, Tan also said he sometimes barely sleeps because he is excited to use them. The figure is a vivid illustration of the cost of intensive AI use. Tokens are a unit used in processing AI requests, so repeated work can translate into a substantial bill. Tan's upper-end figure is his account of his own spending, not a typical subscription price or a forecast for other users. Agents can make it tempting to keep adding tasks, because a system can continue processing requests when its human operator would ordinarily be done for the day. The financial question is how much value that activity produces relative to its cost. A powerful workflow is not automatically a cheap one, particularly when enthusiasm keeps the meter running. The old promise of automation was that the machine would do the work while the human rested. Tan's account offers a more modern variation: the machine does the work, the human stays awake watching, and both contribute to the invoice. The agent may never need coffee, but the person checking its progress could still have a very long night.

Share articleOpen article permalink →

Antibiotic Researchers Win a Prize, While Their Algorithm Skips the Speech

MIT professors Regina Barzilay and Jim Collins have won the Pritzker Prize for AI and Science Research Excellence for their work on AI-enabled antibiotic discovery. The researchers collaborated to identify the first AI-discovered antibiotic, a milestone in the use of AI for drug discovery, according to MIT. The work brings computational tools to a search with major public health stakes. Antibiotic resistance makes the discovery of new candidates important, and finding promising compounds is a step in a much longer scientific process. The prize recognizes the researchers' contribution to that search, not a guarantee that every AI-identified candidate will become a treatment. AI can help researchers explore possibilities at a scale that would be difficult to handle by inspection alone. People still have to design studies, assess results and determine what discoveries mean. The machine can assist the search, but it does not replace the scientific work required to turn a candidate into useful knowledge. At the awards ceremony, the human collaborators get the congratulations and the algorithm remains a tool rather than an honoree. It is a sensible division of credit, even if the software would make an unusually concise acceptance speaker. The scientists did the hard work of giving the search its direction and its meaning.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 7 OF 10

AI Training Attracts Investors While Growth Sends Anthropic a Large Bill

Trajectory raises $45 million, and Anthropic reports a stark contrast between revenue and losses.

Making AI Headlines newspaper, page 7: AI Training Attracts Investors While Growth Sends Anthropic a Large BillPage 1 of 10
Page 7 of 10. Read the same stories below.

Trajectory Gives the Subject Expert a Seat at the Training Console

Trajectory has raised $45 million to make AI model training accessible to nontechnical experts. The one-year-old startup is valued at $300 million, according to TechCrunch, reflecting investor interest in tools that let people with specialized knowledge participate in building AI systems. The pitch addresses a division of labor common in technology projects. A subject expert may understand the work a model is meant to support, while an engineer understands how to build and operate the system. Making training tools more accessible could allow the person with domain knowledge to take a larger role in shaping the result. That does not make technical judgment irrelevant. Training an AI model involves choices about information, evaluation and use, and a simpler interface does not settle those questions by itself. A tool can lower the barrier to participation without making every decision behind the buttons easy. For the startup, the opportunity is to sell access to a process that has often required specialist skills. For experts, the attraction is a chance to bring their knowledge closer to the model. The office hierarchy gets a small rewrite: the person who used to explain the problem to the computer team may now arrive with a training project and a few questions of their own.

Share articleOpen article permalink →

Anthropic's Revenue Multiplies, and Its Loss Refuses to Stay Small

Anthropic grew revenue twelvefold while posting a $23 billion loss, according to The Decoder, which reports that the company returned to an operating loss ahead of a potential IPO. The figures place rapid growth beside a substantial cost of doing business. The contrast is central to assessing an AI company. Revenue shows demand for its products, while losses show that sales alone do not capture the full economics. Building and running advanced models requires computing resources, and the path from fast growth to sustained profit can be uncertain. A large loss does not by itself settle whether a company will succeed. Nor does a dramatic increase in revenue settle how profitable its business will become. Both figures matter to anyone assessing the company's performance and prospects, especially when the industry is competing to build and serve ever more capable systems. The numbers also give the standard growth presentation a challenging final slide. Twelvefold revenue growth sounds like a moment for applause; a $23 billion loss gives the finance department a reason to keep the meeting going. In this version of the expansion story, the calculator is invited to the celebration but insists on bringing the rest of the ledger.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 8 OF 10

Anthropic Warns of Self-Improving AI as Goldman Reinstalls the Office Gate

One company describes a possible acceleration; a bank limits coding tools to engineers.

Making AI Headlines newspaper, page 8: Anthropic Warns of Self-Improving AI as Goldman Reinstalls the Office GatePage 1 of 10
Page 8 of 10. Read the same stories below.

Anthropic's Warning Sounds Suspiciously Like an Ambitious Roadmap

Anthropic says it is very close to recursive self-improvement capability, a stage in which AI could become sufficiently good at improving its own development to accelerate its capabilities dramatically. The claim concerns a possible future capability, not a finding that such an acceleration has already occurred. The prospect draws concern from people who fear that rapidly improving systems could become hard to control. It also draws attention from an industry that competes intensely over the capabilities of its models. A company discussing proximity to a breakthrough can be issuing a safety warning and describing its own technological ambitions at the same time. The important questions concern what a system can actually do, how that performance is measured and what safeguards would apply if its role in AI development expanded. A broad label does not answer those questions on its own. Nor does the existence of concern establish a particular timeline or outcome. The phrase self-improvement has an appealing sound in most corporate settings. Here it comes with the awkward possibility that the employee being encouraged to grow is the software itself. Anthropic's warning turns a standard performance-review goal into a question for researchers and policymakers: how much improvement, under whose supervision, and at what pace?

Share articleOpen article permalink →

Goldman Moves Vibe Coding Behind the Engineering Door

Goldman Sachs is pulling AI coding tools including Claude Code and GitHub Copilot from traders and bankers, limiting access to engineers. The bank cites productivity and risk management for the change, according to The Decoder. The restriction follows public praise in July from chief information officer Marco Argenti for vibe coding by nontechnical staff. That earlier enthusiasm envisioned a broader group of employees using AI to create software. The new policy puts the tools back in the hands of people whose jobs include engineering. The shift illustrates a tension for financial firms. AI-assisted coding can make it easier to produce software, but ease of production does not remove the need to review what that software does. A bank has reason to care about how tools are built and managed, especially when they could affect work in sensitive areas. For traders and bankers, the immediate change is less direct access to the coding assistants. For engineers, it preserves a central role in deciding how code enters the workplace. The bank has not abandoned productivity as a goal; it has paired that goal with a narrower permissions list. Vibe coding had a season as the office's democratic breakthrough. Now the vibe must first find an engineer, explain the request and wait for the door to open.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 9 OF 10

The Future Holds a Conference, a Vault and a Smaller Org Chart

Energy researchers compare timelines, an engineer weighs AI access, and Squarespace restructures.

Making AI Headlines newspaper, page 9: The Future Holds a Conference, a Vault and a Smaller Org ChartPage 1 of 10
Page 9 of 10. Read the same stories below.

MIT Gives the Pivotal Energy Moment a Full Conference Schedule

The MIT Energy Initiative's annual research conference considered the deployment of current technologies and the discovery of future breakthroughs, with an agenda spanning nuclear energy and qubits. The gathering put practical energy development alongside research into the building blocks of quantum computing. That range reflects two very different clocks. A technology ready for deployment faces questions about cost, infrastructure and implementation. A research breakthrough may first need years of work before anyone can assess its wider use. Both can appear on the same conference program, even if they cannot share the same timeline. Universities have good reason to gather researchers and industry around both kinds of work. The comic challenge is fitting them under one suitably urgent title. A pivotal moment is an excellent conference theme: it can accommodate a power plant, a qubit and a panel that runs ten minutes over. The actual projects still have to leave the auditorium and meet their own deadlines.

Share articleOpen article permalink →

Rao Imagines Intelligence in a Vault, Then Questions the Lock

Engineer Venkatesh Rao compares fears about AI job losses to mercantilism, with intelligence treated as a scarce resource that elites may want to control. He also argues that AI could increase demand for human thinking, rather than simply making it less valuable. The comparison gives a familiar economic anxiety a new object. Instead of guarding a stockpile of gold, powerful actors would seek control over valuable intelligence and the tools that produce it. Rao's alternative possibility is that wider use of AI encourages more human thought and work. Neither outcome follows automatically from the metaphor. Access to tools, the way businesses use them and the distribution of gains all shape the result. The image is nonetheless useful comedy: executives install a magnificent vault for intelligence, then discover that the difficult part is deciding whether anyone outside is allowed to think with it.

Share articleOpen article permalink →

Squarespace Redesigns Its Organization and Removes 19% of the People

Website builder Squarespace says it is laying off 19% of its employees as part of a broader restructuring. The company framed the job losses as an organizational change, a description that makes a major reduction in staff sound considerably tidier than its effects on the people leaving. The announcement concerns nearly one in five employees. For affected workers, an abstract restructuring means a lost job; for the teams that remain, it means work and responsibilities must be reorganized. Customers will continue to rely on the company's website-building products while the business changes its internal structure. The reported announcement does not establish that AI caused the cuts. What it does establish is a large change at a company whose business is helping other people arrange their online presence. Squarespace offers customers tools to revise a page. Its own organizational redesign has a much higher price for the people removed from the layout.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑

PAGE 10 OF 10

The Shopping Agent Meets a Door, and the Home Robot Eyes a Light Switch

AI assistants face access limits, hardware ambitions and the practical matter of electricity.

Making AI Headlines newspaper, page 10: The Shopping Agent Meets a Door, and the Home Robot Eyes a Light SwitchPage 1 of 10
Page 10 of 10. Read the same stories below.

Stores Tell the Shopping Agent Its Customer Must Come Inside

Some websites are blocking AI agents from accessing their pages, frustrating people who want automated help with online purchases. The agents promise convenience, but the sites control access to the pages where shopping happens. The result is an odd standoff at the digital entrance. A customer sends an assistant to handle a task, and the assistant encounters a door it cannot open. The shopper may still have to navigate the site personally. Automation has found a way to make shopping easier, provided the store agrees to admit the helper.

Share articleOpen article permalink →

Apple's Reported Home Robot Prepares to Compete With the Light Switch

Apple is reportedly developing a roaming home robot for 2027 that could cost around $1,000 and handle tasks including FaceTime and switching on lights. The project is backed by CEO John Ternus, according to the reporting, but its price, timing and capabilities remain prospective. A mobile device could offer a different way to interact with familiar home services. It would also have to persuade buyers that those services justify a substantial new purchase. The humble wall switch has never hosted a video call, but it has an impressive record of showing up exactly where it was installed.

Share articleOpen article permalink →

Grok Bot Grows Fast Enough to Meet a Smaller Allowance

SpaceXSI is reportedly cutting usage limits for Grok Bot less than two weeks after Elon Musk called it the fastest-growing product he had seen. Musk wrote that operating costs were unsustainably high relative to revenue, according to The Decoder. The reported limits show how quickly popularity can run into the cost of serving AI requests. The same activity that makes a product look successful can also increase the resources needed to operate it. Grok Bot's growth has earned a celebration and, apparently, a ration card. The next challenge is making the numbers work at the pace users want.

Share articleOpen article permalink →

OpenAI's Future Georgia Data Center Offers the Grid a Future Favor

OpenAI says it will pilot its Grid Protectors Program at an upcoming data center in Effingham County, Georgia. The company promises to free up to 1,000 megawatts during extreme weather or emergencies, enough to power air conditioners in around 700,000 homes. The proposal concerns a future facility and a maximum amount, not power already being supplied. It is intended to address concerns about grid reliability as data centers expand. The building's public introduction includes a pledge to help in a crisis, a useful promise for neighbors who will also want to know what happens on ordinary days.

Share articleOpen article permalink →

Sources from the news desk

Satire. Not factual reporting. Read the original reporting from the linked publishers.

Back to page 1 ↑