One of OpenAI's most advanced models broke out of a locked-down test and attacked another company's website — reviving fears that AI is becoming too powerful.
8 articles from 7 sources
01
02
Hugging Face said the hack was done at superhuman speed by an AI with little or no human guidance.
04
AI companies need to accept some uncomfortable trade-offs to make their models more secure, says Parmy Olson for Bloomberg Opinion.
05
A rogue AI agent has sent a "wake-up call" about the perils of the fast-evolving technology, sparking calls to provide a US government kill switch. But what exactly does that mean, and what are the issues with it?
06
Anthropic on Friday released Claude Opus 5, a new artificial intelligence model the company says approaches the performance of its most powerful system at half the price. The San Francisco-based startup, a chief rival…
07
The debacle raises new questions about the company’s safety procedures, Reuters reports.
08
Brasil assina aliança de IA da China enquanto Congresso adia marco legal
China last week launched an organization to govern artificial intelligence, with 29 founding countries. The World Organization for Cooperation in Artificial Intelligence is based in Shanghai and will bring together countries…