Closing the Analyze-to-Design Gap: An Authorization-Bounded Agent for Self-Driving Laboratories
Authors: Yimeng Liu, Eric Isbrandt, Felix Katzenburg, Kevin P. Greenman, Han Hao, Yang Cao, Varinia Bernales, Alán Aspuru-Guzik
Subjects: Artificial Intelligence (cs.AI)
arXiv: 2611.25043 [cs.AI] (submitted 18 Nov 2026)
Abstract
Artificial intelligence (AI) agents are increasingly integrated into self-driving laboratories (SDLs) to plan experiments and interpret results. However, analytical reasoning alone does not make a laboratory autonomous: autonomy in an SDL is achievable only when the agent can be trusted to act on physical hardware unsupervised and knows when to defer to a human. Here, we present an architecture that integrates an agent at the Analyze–Design interface of the Design–Make–Test–Analyze (DMTA) cycle, enabling agent-driven measurement selection and data interpretation while ensuring that state-changing operations are authorized, validated, and auditable.
We demonstrate this architecture on a liquid chromatography–mass spectrometry (LC-MS) instrument in an oxidant-screening campaign for synthesizing an organic dye. Within human-defined campaign limits, the agent submitted measurement jobs, interpreted MS and diode-array detection (DAD) data, flagged failed experiments, and produced an oxidant ranking. This ranking matched that obtained from scripted peak integration, and organic chemists confirmed the agent's spectral interpretations and failure assignments.
In a simulated environment that replays the archived LC-MS data—where the agent chose which wells to inject under varying injection limits—it measured the highest-yielding condition at about its seventh injection, against 14.5 on average for a random order, and identified it in every campaign with a limit of eight or more injections. An agent can therefore close the analysis-to-design gap while its actions stay bounded and verifiable, which is a step toward greater autonomy for SDLs.
Key Contributions
- Architecture for bounded autonomy: An agent integrated at the Analyze–Design interface of the DMTA cycle, with all state-changing operations authorized, validated, and auditable.
- Real-hardware demonstration: Deployment on an LC-MS instrument in an oxidant-screening campaign for organic dye synthesis, including measurement job submission, MS/DAD interpretation, failure flagging, and oxidant ranking.
- Validation against ground truth: Agent-derived oxidant ranking matched scripted peak integration, and chemists confirmed spectral interpretations and failure assignments.
- Simulated injection-selection study: Under varying injection limits, the agent reached the highest-yielding condition at about its seventh injection (vs. 14.5 on average for random order) and identified it in every campaign with a limit of eight or more injections.
Significance
In 2026, as SDLs move from laboratory demonstrations toward routine deployment, the field's central bottleneck is shifting from analytical capability to trustworthy autonomy. This work addresses that shift directly: rather than maximizing agent scope, it constrains agent action with authorization bounds and auditability while still closing the analysis-to-design loop. The result is a practical step toward greater autonomy for self-driving laboratories, where AI agents act on physical hardware unsupervised within defined limits and reliably defer to humans when appropriate.
via ArXiv AI
