1/ Google DeepMind introduced ‘agentic cheminformatics’ in a 2026-11-03 update using Google Antigravity and new Cheminformatics Skills.
2/ The approach is designed to accelerate drug-discovery workflows, and it was said to have completed tasks such as searching thousands of compounds, matched molecular pair analysis, and proposing new macrocycle within one hour.
3/ This announcement shows that AI agents are evolving beyond simple question answering toward handling the repeated steps of scientific work as a unified process.
4/ The key point in cheminformatics and drug discovery is connecting exploration, comparison, and hypothesis generation into a single workflow. [1]
Implications
5/ For materials and scientific researchers, this signals a trend in which complex exploratory tasks are increasingly automated.
6/ Tasks such as molecular candidate search, structure-property relationship comparison, and candidate design proposals may be repeated in shorter time, potentially speeding up early screening and idea generation.
7/ At the same time, result interpretation, data quality, and validation responsibility become even more important.
8/ As such tools become more widely used, people are likely to focus more on experimental validation, objective-function definition, and the incorporation of domain constraints. [1]