Anthropic says AI development may need to slow as systems approach recursive self-improvement. The company proposes global ...

METR, which runs the benchmark measuring how well models can complete long-duration tasks, found that Claude Mythos Preview ...

Sapient researchers trained a 1B reasoning model on just 40B tokens — scoring competitively with 2B-7B models at a fraction ...

Industry chiefs say a looming breakthrough might deliver superintelligence. Safety experts say we’re not ready for it. Read ...

AI models are rapidly improving – so fast that they may soon be able to develop themselves without human involvement. That’s ...

Yuandong Tian, who spent more than a decade as a research scientist director inside Meta’s Fundamental AI Research lab, has ...

Anthropic warns self‑improving AI could outpace human control, urging a slowdown as risks grow and systems begin advancing on ...

Last week, Anthropic boasted that Claude is now writing 80% of Anthropic’s code, which the company presented as a harbinger ...

Claude now authors 80% of Anthropic's production code. The company's new paper maps the path to recursive self-improvement and calls for a global pause mechanism.

Anthropic warns human code review could bottleneck AI progress as Claude writes most code, raising control risks and calls ...