Protein design community balances openness, security

New artificial intelligence tools for biomolecular design enable researchers to build completely new proteins to make protein-based nanomaterials, antibodies, and other drug candidates. The field looks set to keep advancing rapidly, with the amount of computational resources devoted to training biological sequence models doubling about every 4 months. While dazzling, these advances also concern biosecurity experts. "It's very, very hard to separate the benefits from the risks," science policy consultant Sarah Carter tells C&EN. For example, a hypothetical AI tool to identify vaccine targets could also predict vaccine escape, Carter says. In early April, experts gathered at the National Academies of Sciences, Engineering, and Medicine (NASEM) to discuss ways to mitigate the risks that AI poses without delaying scientific progress. "I do think one of the chief risks is being overconservative in this space," Joseph Buccina of the National Security Commission on Emerging Biotechnology said during the meeting. In March,