CERN and Mistral sign statement of intent to develop AI foundation model for physics

The planned model will be pretrained on billions of simulated proton‑proton collisions from the Large Hadron Collider (LHC). CERN stores roughly one exabyte – one million terabytes – of experimental data, and the number of collisions is expected to increase tenfold with the High‑Luminosity LHC, slated to begin around 2030. The model aims to handle multiple tasks such as reconstructing collisions, identifying particles, flagging unexpected signals, and evaluating future detector designs, potentially replacing a chain of specialised algorithms.

Both organisations see the partnership as a way to combine CERN’s data‑intensive expertise with Mistral’s large‑scale model‑training capabilities, creating AI tools that could serve the broader scientific community and strengthen Europe’s technological leadership.