TotalEnergies won’t let go of Mistral AI. Just over a year after their initial alliance, the oil giant and the scale-up are stepping up the pace. This time, the focus turns underground and centers on models capable of detecting oil and gas faster, and deeper.
They formalize a three-year partnership, backed by more than €100 million. The ambition stated is to bring to life a new generation of frontier AI models, tailored for exploration, characterization, and the development of oil and gas reservoirs.
On paper, the equation is straightforward: on one side, nearly 10 petaflops of data amassed by the operator; on the other, nearly a century of experience and know‑how in geosciences and reservoir engineering.
By merging these assets, with a nudge from agentic AI, TotalEnergies hopes to empower its geologists and engineers with tools capable of ingesting massive data volumes to multiply scenarios; whether it’s to uncover new exploration opportunities or to extend the productive life of existing fields.
The two partners will establish a joint scientific laboratory, where TotalEnergies’ subsurface teams will collaborate with Mistral’s researchers.
“Exploration and reservoir engineering are among the areas where artificial intelligence can create the most value for our activities,” argues Patrick Pouyanné, the CEO of TotalEnergies, who also sees it as a way to “contribute to the development of a high-performing European digital ecosystem.”
On the other side, Arthur Mensch, co-founder and CEO of Mistral AI, heralds an AI that is “customizable” and respectful of the intellectual property rights of large groups. A compelling argument in the face of competition from the American giants.
