Industrial Alchemy: Microsoft Agents Replace Years of R&D for Construction Giant

Artificial intelligence is proving its ability to radically accelerate fundamental science. On October 20, 2026, Microsoft and Saint-Gobain unveiled the results of applying the Discovery AI platform in the development of low-carbon cement. Autonomous algorithms analyzed 62,000 patents and 44,000 scientific abstracts in just a few hours—a workload that would have required a year of manual labor from an R&D department.

The system generated ready-made hypotheses for new chemical compounds to reduce the carbon footprint. For the DeepTech macroeconomy, this case illustrates the real monetization of generative models in the physical world (Physical AI). While the B2C market is entertained by chatbots, heavy industry is using AI as a computational collider. Companies capable of integrating such scientific tools into their R&D pipelines will gain an unassailable patent advantage, reducing the cost of primary research to zero.

Source: Microsoft / Saint-Gobain