← Back to Model Beat
6Other·May 22

Measuring LLMs’ ability to develop exploits

Measuring LLMs’ ability to develop exploits Anthropic

Covered by 1 source

Related stories

OtherA new era for AI SearchMay 19 · 2 sourcesOtherWidening the conversation on frontier AI - AnthropicMay 19OtherFast-tracking genetic leads to reverse cellular agingMay 18OtherAn OpenAI model has disproved a central conjecture in discrete geometryMay 20