The A-Lab at Berkeley is closing the loop in the lab by integrating robotic synthesis and characterization with machine-learned data interpretation and AI-driven decision-making to iteratively revise synthesis routes, mirroring parallel efforts like a mobile robotic chemist at the University of Liverpool.

Pre-synthesis screening is advancing, with approaches like Google's GNoME using graph neural networks and density functional theory to identify stable crystal structures, and Microsoft's MatterGen generating new crystals conditioned on target properties to propose plausible candidates.

Commercial platforms are expanding the Design-Make-Test-Analyze loop, and Dunia Innovations reports substantial gains in electrochemical testing and performance through ML-guided design and automated experiments.

Industry validation and investment are accelerating AI-enabled labs, as researchers point to surging scientific output and startups such as Periodic Labs and Lila Sciences raising funding to push autonomous laboratories forward.

Looking ahead, Nature Reviews Chemistry highlights scalability, generalizability, and provenance as core requirements, calling for interoperable data standards, modular hardware, and AI agents that reason under uncertainty with full traceability from measurement to sample.

A necessary correction has been issued: initial novelty claims for A-Lab materials were revised after independent analyses, clarifying that the materials were new to the prediction platform rather than novel to science, and underscoring challenges in automated diffraction analysis and incomplete compound databases.

The field is moving from automation toward autonomy, with autonomous experimentation where multimodal AI models reason about experiments and guide robotic work using approaches like CRESt, large language models, and Bayesian optimization.

Overall, AI and automation are accelerating materials discovery by combining ML with robotic labs, potentially speeding the journey from idea to product by a factor of 10 to 100.