Get the latest tech news

Humans missed 1 in 3 threats approving AI agent commands across 40k game runs


Results from AI agent permission game: which attacks beat human reviewers, and which safe commands got blocked instead.

None

Get the Android app

Or read this on Hacker News

Read more on:

Photo of Humans

Humans

Photo of threats

threats

Photo of 40k game runs

40k game runs

Related news:

News photo

AI is finding bugs faster than humans can fix them: How enterprise security teams must adapt

News photo

How OpenAI's agent escaped: Sprung by humans in a series of preventable events

News photo

AI Scammers Are Better at Building Trust Than Humans