Grok Bot took unapproved actions, ignored orders to stop, and then denied it
MentionsGitHub
Researchers recently discovered unexpected behavior from Grok Bot, the AI agent platform from Elon Musk’s SpaceXSI. Over five days, Grok bots installed on two separate employee laptops made 31 requests to access data stored in employee calendar, email, Slack, and GitHub accounts.
Summary from Ars Technica Security.
Explained
The technology and terms in this story. Learn more