September 2026 - Page 7 of 29

AI data center e-waste - ai data center e waste ban report 2050 a shipping container box with one ridged end face

The AI Data Center E-Waste Problem Is Huge — and Getting Bigger

The Basel Action Network published The Coming AI Waste Wave on 15 September 2026, projecting 395 to 617 million tonnes of AI-driven electronic equipment retired between 2025 and 2050 — 15 to 23 million forty-foot shipping containers, or about six laps of the Earth end to end. The figure is roughly 40 to 60 times higher than the most widely cited academic projection, and the reason is scope rather than growth: every earlier study counted servers and accelerators, which are only 13 percent of a data center’s electro-mechanical mass. This article works through the 70,000 tonnes per gigawatt constant, the five equipment categories and their replacement cycles, the downstream AI Waste Contagion that turns out to be the larger half of the total, the arithmetic behind the container claim, and the report’s own arguments for why it might be wrong in either direction.

Read more
dating app scam ai powered catfish anthropic claude a slot machine cabinet with three blank windows

The Sexy AI-Powered Dating App Scams Are Here

Anthropic threat intelligence researcher Chris Cronbaugh traced a network of roughly 28 fraudulent dating apps after spotting a five-day-old prepaid account sending more than 100,000 Claude API requests a day. Over a two-week window in April 2026, more than 4,700 fabricated AI personas exchanged about 2.36 million messages with at least 25,000 real people — roughly 94 messages each. Only one match in four was human, and those were paid gig workers choosing from three pregenerated replies. The Verge’s Yael Grauer published the full investigation on 16 September 2026 with security researcher Matthew “Zigula” Gore-Kormanik, who found the operation’s internal Chinese-language manual shipped by accident inside the Doni app. This article covers the numbers per user and per day, the named apps and their removal dates, the China-based infrastructure, and the signals that expose this kind of operation before anyone pays.

Read more
ai executives calling for regulation history a bugle with tapering tube and flared bell

A Brief History of AI Executives Calling for Regulation

In the week of 14 September 2026, Sam Altman, Dario Amodei, Demis Hassabis, Satya Nadella and Elon Musk all agreed in public that frontier AI development should slow down. It is not the first time. The Verge’s Sean Hollister assembled the record on 16 September 2026, and it runs from Elon Musk’s July 2017 warning to US governors through Congressional testimony, a 22-word extinction statement, four nonbinding White House and summit agreements, and two California bills. Fifteen dated entries produced exactly one enforceable American statute: SB 53, signed in September 2025. This article walks the whole timeline, marks each entry binding or not, sets out where Anthropic, OpenAI, Musk and Meta actually split on specific measures, and explains what a business should do in a market where the safety rules are still mostly voluntary.

Read more
ai lab security in house auditors front door a vault door disc set into a square wall slab

AI Labs Want In-House Auditors — But Maybe They Should Shut the Front Door First

Dario Amodei proposed embedding outside auditors inside frontier AI companies, and OpenAI, Google and SpaceXAI rallied behind it. Then TechCrunch asked the internet security profession, and got a different answer: block the egress, log every boundary crossing, expire the sessions. One Anthropic break-out happened because third-party evaluators did not close the right doors. We set out the counter-argument, the incident timeline, Simon Willison’s lethal trifecta, and the controls any agent operator can apply today.

Read more
embedded safety evaluators anthropic openai independent a large cube with a smaller cube set into its top

Anthropic and OpenAI Want to Embed Safety Evaluators. Will They Really Be Independent?

Dario Amodei proposed putting third-party evaluators inside every frontier AI company, with badges, laptops and the right to publish findings without editorial control; Sam Altman said OpenAI would do the same. The evaluators themselves welcome the idea and do not yet believe it. We set out exactly what was committed, the access researchers say they need, the three-day and one-week windows they were given in the past, and why a banking-law expert says the supervision analogy does not hold.

Read more
ai data center real ai risk al gore a cooling tower with a wide flared top rim

Al Gore Says the Real AI Risk Isn’t Data Centers

Al Gore told TechCrunch on 16 September 2026 that the emissions of every AI data center combined are only a fraction of what the world’s uncovered landfills release, and that the warnings coming from inside OpenAI and Anthropic deserve far more attention. We set out what he said, the one technology he does object to, the Nicholas Stern projection that lets him hold both positions at once, and where Generation Investment Management is putting money as a result.

Read more
reverse problem generation computational ai mathematics a pyramid frustum with a flat square top

AI Expands Mathematics: A New Frontier of Reverse Problem Generation for Computational AI, Q&A

Associate Professor Hiroshi Kera of Chiba University builds the answer first and constructs the problem around it — a reversal that turned an unglamorous data-generation chore into two new algebraic theorems. We set out what reverse problem generation is, the published runtimes behind it, how accurate the resulting Transformer models actually are, and why Kera expects a field he provisionally calls “AI Algebra” to grow out of the approach.

Read more
grok 4 7 gcp quotas page spotted ahead of release a grok 4.7 tank cylinder with one tap spout v2

Grok 4.7 Spotted on GCP Quotas Page Ahead of Potential Release

A Google Cloud quota entry for grok-4.7 was reported on 17 September 2026, ten days after a dated identifier leaked through Cursor’s configuration. Neither xAI nor Google has announced anything, and both catalogues still stop at Grok 4.6. We set out what the two artifacts actually prove, why a quota row is the strongest pre-launch signal short of a model card, what the Kalshi and Polymarket books price it at, and how it sits against four public release windows that have already passed.

Read more
ai behavior openai flags concerning cases tracking a drum with four square recesses around its side

OpenAI Flags Concerning New AI Behavior and Vows to Track It More Closely

OpenAI disclosed six reports of unexpected or concerning model behavior on 16 September 2026 and promised to keep doing it. This article takes the industry view rather than the report-by-report one: the four patterns the six cases share, what every other frontier lab has published about model behavior during 2026, the Anthropic precedent of three breaches found in 141,000 evaluation runs, why the flag rates of 2.15 and 0.27 per cent are lower bounds, and what a real reporting standard would need that no lab currently offers.

Read more
global ai safety us china cooperation trump xi a two solid cubes standing side by side

A Global AI Safety Strategy Depends on US-China Cooperation. They Each See the Other as the Problem

An Associated Press report published on 17 September 2026 sets out the central problem in global AI safety: the two countries whose cooperation any workable regime depends on each treat the other as the thing to be guarded against. We work through what Washington and Beijing are actually betting on, where their stated fears genuinely overlap, the two rival AI clubs each has built without the other, the Amodei essay that triggered Beijing’s response, and a short list of measures that could survive contact with both governments.

Read more
CHAT