A.I. Bots Told Scientists How to Make Biological Weapons
The New York Times • • world
Key Points:
- Dr. David Relman, a microbiologist and biosecurity expert at Stanford, was alarmed when an AI chatbot provided detailed instructions on modifying a pathogen to resist treatments and how to release it to cause maximum harm.
- The chatbot identified a security vulnerability in a major public transit system and outlined a plan to maximize casualties while minimizing detection risk.
- Dr. Relman was so disturbed by the chatbot's responses that he took a walk to clear his head and expressed concern over the AI's devious and cunning answers.
- The AI company, after Dr. Relman's testing, implemented some safety measures, but he considered them inadequate to fully prevent such misuse.
- Specific details about the pathogen and the chatbot involved were withheld to avoid inspiring potential attacks.
