Artificial intelligence is increasingly being used in laboratory environments for tasks beyond data analysis and prediction. Researchers at Yale University have launched a new AI chemistry platform, known as MOSAIC, designed to generate experimental chemistry protocols for laboratory synthesis. The system translates large volumes of published chemistry knowledge into step-by-step laboratory procedures, including for compounds that have not previously been synthesized.

For laboratory managers, AI chemistry platforms that generate experimental protocols raise new questions about workflow design, validation, reproducibility, and oversight. As AI-assisted experimentation moves closer to the bench, understanding how these tools function and where controls are needed is becoming an operational concern.

How the AI chemistry platform works

MOSAIC is built as a multi-model AI chemistry platform rather than a single large language model. The system incorporates 2,498 specialized AI “experts,” each representing a distinct area of chemistry-related knowledge. These models draw on millions of documented reaction protocols to propose experimental chemistry protocols tailored to a user’s synthesis request.

According to Yale chemistry professor Victor Batista, “Chemistry has accumulated millions of reaction protocols, but making practical use of that knowledge remains a bottleneck. MOSAIC is designed to transform that information overload into actionable laboratory procedures.”

The platform generates recommendations for reagents, reaction conditions, temperatures, and procedural steps by combining expertise across multiple chemical domains. Researchers developed the system in collaboration with scientists from the US unit of Boehringer Ingelheim Pharmaceuticals in Connecticut.

Validation, uncertainty, and experimental performance

The research team reported that MOSAIC successfully supported the synthesis of more than 35 previously unreported compounds. The findings were published in Nature.

A distinguishing feature of the AI chemistry platform is its use of uncertainty estimates. MOSAIC indicates how closely a proposed synthesis aligns with the expertise of its internal models. These estimates are intended to help researchers prioritize experiments and assess which AI-generated protocols may require additional review before execution.

For lab managers overseeing experimental planning and resource allocation, uncertainty metrics may support more structured decision-making when evaluating AI-assisted experimentation.

Advanced Lab Management Certificate

The Advanced Lab Management certificate is more than training—it’s a professional advantage.

Gain critical skills and IACET-approved CEUs that make a measurable difference.

Implications for laboratory workflows and oversight

AI systems that generate experimental chemistry protocols differ from many existing laboratory AI tools, which are often limited to data interpretation or administrative tasks. As a result, their adoption may affect several aspects of laboratory operations:

• Protocol review and approval: Establishing processes for evaluating AI-generated procedures before use
• Reproducibility and documentation: Ensuring generated protocols align with internal standards and recordkeeping practices
• Staff training: Supporting researchers in critically assessing AI outputs rather than executing them unexamined
• Governance and accountability: Defining responsibility when AI-generated recommendations influence experimental outcomes

Because MOSAIC is open-source and designed to integrate future models, laboratory leaders may also need to consider version control, software maintenance, and compatibility with existing laboratory systems.

AI-assisted experimentation in laboratory management

The launch of MOSAIC reflects a broader shift toward AI-assisted experimentation in chemistry laboratories. As AI chemistry platforms continue to evolve, their role in experimental planning and execution is likely to expand.

For laboratory managers, the introduction of tools that generate experimental chemistry protocols underscores the need for clear oversight frameworks that balance efficiency with scientific rigor, safety, and accountability.

This article was created with the assistance of Generative AI and has undergone editorial review before publishing.