On October 16, 2026, Google Research outlined three new projects aimed at grounding AI agents in physical laboratory robotics to drive autonomous wet-lab experiments.
Bridging AI and Lab Reality
Google Research announced initiatives to connect high-level AI reasoning with physical wet-lab robotics, focusing on closed-loop molecular design and safe experimentation. The researchers introduced the VERified Infrastructure for Discovery Laboratories (VERID) framework to serve as a semantic digital twin of the laboratory workcell.
Built on top of the open-source PyLabRobot framework, VERID offers live 3D state tracking, fast virtual dry runs, hardware safety checks, and synchronized video and action logs. Google said these capabilities help agents plan, simulate, and execute liquid-handling protocols while avoiding physical collisions or empty wells.
Aptamer and Protein Discoveries
In tests using protein-binding DNA aptamers across three design-test-iteration cycles on myoglobin and alpha-thrombin, Google reported that AI-driven rational and de novo design raised confirmed on-target hit rates to 33.5% and 42.1% respectively. The workflows enabled agents to identify wet-lab failure modes and iteratively revise their design rules.
For protein binder design targeting fibroblast activation protein alpha (FAP), Google applied surgical epitope excision to accelerate structure prediction 8.6x. The human-agent hybrid pipeline produced top designs with single-digit nanomolar affinity in just a few experimental runs.
Key facts and where they come from
- Google Research announced the VERID framework on October 16, 2026, to bridge AI reasoning and wet-lab physical reality.
First, we present the VERified Infrastructure for Discovery Laboratories (VERID) framework to bridge the gap between the high-level reasoning of AI agents
- VERID is built on top of the open-source PyLabRobot framework.
Built on top of the open-source PyLabRobot framework, VERID provides four core capabilities:
- In aptamer discovery, confirmed on-target hit rates reached 33.5% for myoglobin and 42.1% for alpha-thrombin.
raised confirmed on-target hit rates as can be seen in the image below (to 33.5% and 42.1% for myoglobin and α-thrombin, respectively)
- Surgical epitope excision accelerated structure prediction 8.6x for FAP binder design.
surgical epitope excision accelerates structure prediction 8.6x
