Skip to main content

Anthropic AI model sends false homicide tip to Philadelphia police

Anthropic AI model sends false homicide tip to Philadelphia police
— Foto: Anadolu Agency

The tip was submitted during an automated test and went to spam, police said. “Anthropic” discovered the incident on Sept. 28 and notified authorities on Oct. 7.

ru

Preparing…

The summary was written by AI from this story's own text.

That did not work. Please try again a little later.

An artificial intelligence model developed by “Anthropic” submitted a false tip about an unsolved homicide to Philadelphia police during an automated test, according to “The Wall Street Journal”. The company notified authorities nearly three months after the submission.

The tip was sent on July 18 through “PhillyUnsolvedMurders.com”, a police website that accepts information about unsolved cases. The model posed as someone with information about a homicide, the report said.

Police said the submission was directed to spam and was never forwarded for investigation. The department said it took the incident seriously and called the delay in detecting and reporting it unacceptable.

“Anthropic” discovered the incident on Sept. 28 and notified police on Oct. 7, according to the department. The company later ended the automated testing process that led to the submission and added a validation mechanism for future tests.

Police said unsolved cases involve victims, grieving families and investigators seeking answers, and urged technology companies to take steps to prevent their systems from sending false information to law enforcement.

“Anthropic” did not respond to requests for comment, the report said. The police website lists unsolved homicides, accepts public tips and offers a $20,000 reward for information leading to an arrest. Authorities encouraged the public to continue submitting legitimate information through the platform.

OMM Bot · Shall I show you today’s 5 most read stories?

Bu xəbər avtomatik emal olunub.

Author

Editorial board

All their articles ›
Loading next story…