AI data centers are buying up the world's memory chips. That's raising phone, laptop and console prices and killing budget models, likely through 2027.

OpenAI and Meta warn U.S. judicial panel about deepfake evidence

Bank of America hired 1,000 contractors to patch software flaws exposed by Anthropic’s Mythos, offering a rare look at costs involved in defending against AI.

Disney bid for Character.AI's code, not its chats. Beijing's veto doubled Manus's price. Five stories on why AI control now matters more than capability.

NEW YORK – New York Attorney General Letitia James and a bipartisan coalition of 39 other attorneys general today warned that Character.AI should not be sold

Supermarkets in France and the Netherlands are dealing with the aftermath of a cyberattack that disrupted refrigeration systems

A field guide to the texts that arrive every fall between the candidates, the tree guy and the package that doesn't exist.

Political texts, tree services, fake package alerts and Black Friday blasts. A funny field guide to fall text season, plus the only replies that work.

The following is the full transcript of the interview with Shyam Sankar, chief technology officer of Palantir, a portion of which aired on "Face the Nation with Margaret Brennan" on Oct. 11, 2026.

Who grades the graders? Five AI checks that came up short this week Every story in this issue turns on a check of some kind. A panel scoring a video, a lab publishing a safety test, a grader scoring a...

Posting every day makes someone a creator. It doesn't make them an influencer, and advising for free isn't a business.

Samsung's $80B quarter, Apple's iPhone 18 Pro order cuts and TSMC's Q4 guide: how the AI memory crunch reaches your budget the week of Oct 12.

The comforting version of the AI-and-work story goes like this. The tools take the routine tasks.

OpenAI publishes new results on open problems in mathematics from an internal frontier model and shares Lean proof formalizations and research details on GitHub.

A 75% price cut and a vanishing $100 phone are the same story The cheapest useful Claude just got about 75 percent cheaper, and by one outside estimate its maker is already underwater on its heaviest ...

A wave of solidarity posts around the Cornell case points to a bigger question most people haven't had to think about: what can a pseudonym still protect when almost anything can be searched?

Hospitals’ increasing use of artificial intelligence for patient coding has increased the frequency of inpatient stays classified as medically complex—boosting the bills sent to payers despite no a | ...

Ex-AI researchers warn NYC’s AI hearing that humanity may lose control. OpenAI, Anthropic, Google and Meta could not quantify the worst-case risk.

A deputy ran a Flock search because of a California plate. A federal judge threw out the evidence. Flock expects the ruling to be overturned.

Issue #27 | October 6, 2026

The FTC, a lawsuit with no damages claim, and a NYC Council hearing all target the same AI agent breach. What it means for AI accountability in 2026.

Some of fall’s biggest films take on the tech titans who run our world, including Sam Altman, Mark Zuckerberg and Elon Musk.

SB 947 bans employers from letting AI fire someone outright. It doesn't require the human in the loop to have the time, power, or information to disagree.

California's SB 947 bans AI-only firings, but requires no time, data access, or protection for the human who has to review the decision.

Tavus unveils Griffin, the first AI model to pass a video Turing test, with 48% of participants believing they chatted with a real human.

Two courts are deciding who's liable when an AI agent acts alone, Delta's earnings test AI layoffs, and a quiet FCC rule nobody's covering yet.

3 OpenAI researchers fired after a probe into sensitive data shared with an AI safety nonprofit. Read what the investigation found

Something like this has happened before, more than once in a single working lifetime for anyone over forty. The personal computer, then the internet, then the smartphone, then cloud software.

An investigation that predates the incident everyone will blame it on opened into public view this week. The FTC confirmed it has examined Anthropic, OpenAI, and the nonprofit evaluator METR since thi...

Apple has paused a proposal to eliminate roughly 5,000 AppleCare customer-service positions through AI, even as automated support tools continue handling some customer questions and troubleshooting.

Dismissing growing concerns about AI safety, Trump has argued that “Artificial Intelligence” was a "very ineloquent" name for the technology and makes it sound "fake."

The new agents will be available to millions of Robinhood users, who can choose between OpenAI and Anthropic models.

The standard picture of military transition looks like arithmetic. Take a service member close to their exit date, run them through a week or two of resume workshops, hand them a donated suit for inte...

THIS WEEK, IN 60 SECONDS CAPITAL: Firecrawl raised $75 million to build a licensed data marketplace built specifically for AI agents, not for training models. RESEARCH: Anthropic ran 201 employees' AI...

ShinyHunters, a cyber-extortion outfit, said it hacked into the FBI’s jobs portal and used that access to steal a major tranche of sensitive files.

Peter Max died on September 14, 2026, at 88, and most of the obituaries led with the color and the ubiquity: the posters, the album covers, the psychedelic pop-art eye that made his work some of the m...

We obviously are not in any way looking to stifle innovation,” City Council Speaker Julie Menin told Fortune.

The auto-shutdown system that should have killed the training run failed. Monitoring flagged the breach in 15 minutes. The model kept running for two and a half hours after that.

OpenAI paused its own frontier training Friday, the second time in three months, and this time the trigger wasn't an internal slip. Autonomous agents built on its models spent part of the summer probi...