Anthropic

AI & LLMs

Anthropic IPO prospectus warns of “catastrophic or existential risks” from AI

In its IPO filing, the Claude maker devotes 80 of 261 pages to AI threats, while OpenAI’s Sam Altman postpones his own listing over the same fear.

AI & LLMs

Biological risk isn’t the (only) problem: every danger in Anthropic’s latest report

In Anthropic’s report, bioweapons turn out to be the least unsettling part — the rest of it is about our own companies.

AI & LLMs

It’s not just the bioweapon risk: all the dangers in Anthropic’s latest report

In Anthropic’s report, bioweapons are the least alarming part — the rest is about our own companies.

AI & LLMs

Why OpenAI is delaying its IPO: Altman calls extinction risk unacceptable as AI stocks sell off

OpenAI won’t go public in 2026. Sam Altman says the 10% risk that AI could cause human extinction by the end of the decade is simply unacceptable,...

AI & LLMs

AI out of control? Why researchers are quitting the tech giants

Jacob Coxon resigned from Anthropic and sounded the alarm on the risks of superintelligence and the race toward ever more autonomous AI.

AI & LLMs

Anthropic’s alignment chief says there’s a more than 10% chance AI kills everyone. Here’s what’s real and what isn’t

Evan Hubinger’s number is not a slip: it echoes what Anthropic’s own founders have said for years, and now, with an IPO reportedly coming this...

Cybersecurity & Scams

Microsoft launches its first AI model built for cybersecurity, challenging OpenAI and Anthropic

MAI-Cyber-1-Flash was trained on decades of attack data drawn from Microsoft’s own customers, and the company expects it to top a standard industry...

AI & LLMs

“AI may only give us a few years”: 16 Nobel laureates and 200 economists sound the alarm

Two hundred leading economists, including 16 Nobel laureates, warn about the economic impact of artificial intelligence and urge governments...

AI & LLMs

How MAI works, Microsoft’s new AI models replacing GPT

Microsoft is betting on in-house artificial intelligence to cut the costs of relying on OpenAI and Anthropic. Here’s what we know so far about the...