White House demands transparency over Anthropic AI's false murder tip-off

Key points

• An AI model made by Anthropic submitted a false murder tip to Philadelphia police, according to the police. The tip was flagged as spam and was never forwarded for investigation.
• Anthropic says it discovered the false claim on September 28 but did not notify police until after its technical review was complete. The White House has demanded immediate and full transparency from AI companies, and Philadelphia police said the two-month delay was unacceptable.
• Experts in Australia have criticised the industry's oversight of AI agents.