Google DeepMind unveils SynthID Bio to watermark AI-generated proteins while preserving function

Google DeepMind described SynthID Bio as a family of watermarking methods designed specifically for synthetic biology to strengthen biosecurity and scientific integrity. The approach embeds an imperceptible signature directly into the biological code of proteins, ensuring the watermark remains verifiable not only in digital models but also in synthesized, physical proteins.

The team adapted the watermarking method depending on the type of data: guiding amino acid choices in sequences and adjusting atomic coordinates in predicted 3D structures to create a reliable detection signal. In experiments, these adjustments did not compromise the protein’s biological function, which the authors note is essential for therapeutic efficacy and advancing scientific research.

Wet-lab testing focused on protein binders—molecules engineered to selectively attach to other proteins—using AlphaProteo for binder design and a SynthID Bio-enabled version of ProteinMPNN for sequence generation. Across three targets—VEGF-A, the SARS-CoV-2 spike protein receptor-binding domain, and PD-L1—the watermarked designs matched the hit rate, binding affinity, and natural sequence diversity of unwatermarked versions, producing the first biologically functional and watermarked protein binders.

For protein folding, SynthID Bio fine-tunes a small part of AlphaFold 3’s diffusion network to watermark predicted 3D coordinates directly into the model’s weights. This ensures the predicted structure inherently carries a detectable signature regardless of who runs the model. The method preserved AlphaFold 3’s prediction accuracy while achieving near-perfect detectability and maintaining key structural feature distributions even under digital noise or minor coordinate changes.

The authors position SynthID Bio as a verification layer within a broader biosecurity strategy, complementing model-level mitigations and customer vetting. Sarah Carter, a biosecurity policy expert who reviewed the work, said SynthID Bio is 'an important piece of the puzzle for tracking the provenance of biological designs.'