Google Develops AI-Designed Protein Watermarking Method to Counter Biosafety Threats

Techhub News reports that Google's research team has developed a method to add watermarks to AI-designed proteins. This initiative aims to address biosafety risks, as current software for identifying potentially hazardous proteins cannot effectively distinguish those generated by AI.

The watermarking technology helps differentiate AI-designed proteins from natural ones, enhancing regulatory and traceability capabilities. This method is seen as a leading initiative in the biotech field to address potential threats posed by AI. While AI tools have succeeded in designing enzymes that aid in digesting plastic or blocking toxic proteins, they could also be used to create toxins or alter viral protein behavior. (Ars Technica AI)