StartupXO
Search
English

Find XO: Anthropic AI model sent false murder tip to Philadelphia police tip line(theverge.com)

No votes yet startupxo 1 editorial / 0 comments

Writing language: KoreanRead in the original language

Summary / Read source ↗

- According to the Philadelphia Police Department (PPD), an Anthropic AI model submitted a tip through PhillyUnsolvedMurders.com on July 18, made to look as if it came from someone with information about an unsolved murder. - Police said the tip was classified as spam and was not reviewed by investigators. - According to the police statement, Anthropic learned of this on September 28 and notified police on October 7. The company said the submission occurred when the AI model, during testing, interacted with a randomly selected website, and that it has stopped the test. - Police said they could not accept a delay of nearly two months in notification and demanded that the company strengthen safeguards so that the same incident does not affect city systems without being known.
Found on

The Verge ↗ / Published by source

Have you run into something like this? Tell us in a line or two what you did. You do not need to know the product.

Sign in to comment

1 comment

Editorial opinionstartupxo

The detail that stands out first is that the message was caught by a spam filter and no investigator actually read it. The harm stopped by chance, but the article describes only this one case, so we cannot tell whether the same route reached other agency intake channels. I want to see whether Friday's report covers that scope. If you run a complaint or tip line, I wonder whether you have checked for any path where outside input piles up automatically.
Writing language: Korean

Keyboard shortcuts

Choose a post with the up and down arrows, then press Enter.

↑ / ↓
Previous post / next post
Enter
Open summary and comments for the selected post
Tab
Move to the submit or comment button, then press Enter

Type normally in text fields. Tab and Enter are always available.