N-Day-Bench – Can LLMs find real vulnerabilities in real codebases?

Related Stories

• Regulators in Europe were largely left out of the loop as Anthropic limited the release of its latest model Mythos to select companies and organizations
• Sources: Anthropic largely left European regulators out of the loop as it limited Mythos's release to select companies and organizations; the UK AISI tested it
• The Guardian view on AI politics: US datacentre protests are a warning to big tech - In both Republican and Democratic states, scepticism and hostility towards an unregulated construction boom is growing
• OpenAI has bought AI personal finance startup Hiro
• OpenAI acquires personal finance startup Hiro Finance for an undisclosed sum; Hiro stops new signups, will shut down on April 20, and delete all data on May 13