Efforts are being made to increase biosecurity in light of the possibility of manufacturing dangerous proteins with AI
Artificial intelligence (AI)-assisted protein engineering is enabling advances in the design of new molecules, but it also poses biosafety challenges related to the potential production of harmful or dangerous proteins. Some of these threats, whether deliberate or accidental, may not be detected by current control tools. An international team has analyzed the situation and developed software patches to improve their identification, although they acknowledge that it remains incomplete. The authors of the study, published in the journal Science, warn that some of the data and code should not be published in a public repository due to its potential misuse.