AI designs for dangerous DNA can slip past biosecurity measures, study shows

AI designs for dangerous DNA can slip past biosecurity measures, study shows

Companies that make DNA for science labs screen out any requests for dangerous bits of genetic material. But a new study shows how AI could help malevolent actors get the stuff anyway.(Image credit: Malte Mueller/fStoap)

Microsoft researchers say AI models can be used to design toxins that evade biosecurity systems used to screen DNA orders for potential biothreats

Antonio Regalado / MIT Technology Review: Microsoft researchers say AI models can be used to design toxins that evade biosecurity systems used to screen DNA orders for potential biothreats  —  A team at Microsoft says it used artificial intelligence to discover a “zero day” vulnerability in the biosecurity systems used to prevent the misuse of DNA.