Blab AI
Fermilabs Fermi Data Platform Powers DOEs Genesis Mission AI Cloud
When a particle‑physics laboratory partners with a national AI initiative, the result is a data engine that could rewrite the pace of discovery.
Fermilab’s Fermi Data Platform (FDP) has been chosen as a core partner for the U.S. Department of Energy’s (DOE) Genesis Mission, a $37 billion effort announced in November 2025 to accelerate scientific breakthroughs through artificial intelligence. The partnership gives the American Science Cloud—Genesis’s cloud‑based research ecosystem—petabytes of secure storage, data‑access tools, and the institutional expertise needed to enable researchers across the country to apply AI to large scientific datasets.
Genesis brings together 17 DOE national laboratories, universities, and industry partners with the explicit goal of cutting the time between a scientific question and a meaningful answer. The program will automate literature searches, preliminary simulations, and result filtering so that scientists can focus on interpretation rather than logistics. “Give me the 10 most promising materials for batteries – the system does a literature search, runs some simulations to verify, narrows down that list, and presents it as an answer for further research,” said DOE official Gutsche.
Fermilab’s reputation for handling massive data streams makes it a natural fit. The FDP is built on thousands of hard drives and already supports data from CERN’s Large Hadron Collider CMS experiment, the Short Baseline Neutrino program, quantum‑information research, microelectronics development, and advanced theory work. It is also preparing to host data for the upcoming Deep Underground Neutrino Experiment.
According to FDP’s lead partner, Chin Guok, “Data is the common denominator behind major scientific endeavors, and AI is fundamentally data‑driven. To train and run AI models, you need large volumes of data. Fermi Data Platform can support AI training and inference on large scientific datasets.” The platform’s architecture prioritizes repeatable, high‑throughput access that AI‑supported research demands.
The American Science Cloud will layer the FDP’s storage with AI services that let researchers describe a need and have AI tools tap into national‑lab resources—including supercomputers, scientific datasets, and simulation capabilities—without manual coordination. The cloud’s design is intended to be integrated, so a scientist can request a task and receive a refined set of results, such as a narrowed list of candidate battery materials.
Fermilab’s involvement was enabled by the laboratory’s recent management transition in 2025, which brought the Fermi Forward Discovery Group (FFDG) to oversee operations. The laboratory’s expertise in data‑management and its existing Tier‑1 role in the Worldwide LHC Computing Grid positioned it as a natural fit for Genesis’s data needs.
A key challenge for AI in science is that raw data from instruments often lacks the structure and metadata required for machine‑learning models. The FDP bridges that gap by storing datasets in an AI‑ready format and providing tools for data discovery and access.
Genesis’s broader strategy includes deploying AI on 26 national‑science and technology challenges with the full $37 billion budget. The American Science Cloud is one of the mission’s flagship platforms, intended to be a shared infrastructure that supports researchers across disciplines—from high‑energy physics to materials science to fusion‑energy research.
While the FDP’s exact storage capacity is described only as petabytes, the platform’s scale is sufficient to hold the large datasets generated by CMS, neutrino experiments, and quantum research. Its secure, large‑scale infrastructure is expected to accelerate AI‑driven discoveries by making data readily available for training and inference.
The partnership is still in its early stages. Fermilab has begun offering data‑storage and data‑access tools engineered for AI‑supported research, and the American Science Cloud is developing the AI services that will interface with the FDP. Next steps involve integrating the platform into the cloud’s workflow, testing AI pipelines on real scientific data, and expanding the data catalog to include additional experiments.
In summary, Fermilab’s Fermi Data Platform is a key component of the DOE’s Genesis Mission, providing the storage, tools, and expertise needed to make large scientific datasets AI‑ready. The partnership aims to speed up the scientific process by automating data discovery, simulation, and result filtering, thereby enabling researchers to focus on insights rather than logistics.
Fermilab’s Fermi Data Platform (FDP) has been chosen as a core partner for the U.S. Department of Energy’s (DOE) Genesis Mission, a $37 billion effort announced in November 2025 to accelerate scientific breakthroughs through artificial intelligence. The partnership gives the American Science Cloud—Genesis’s cloud‑based research ecosystem—petabytes of secure storage, data‑access tools, and the institutional expertise needed to enable researchers across the country to apply AI to large scientific datasets.
Genesis brings together 17 DOE national laboratories, universities, and industry partners with the explicit goal of cutting the time between a scientific question and a meaningful answer. The program will automate literature searches, preliminary simulations, and result filtering so that scientists can focus on interpretation rather than logistics. “Give me the 10 most promising materials for batteries – the system does a literature search, runs some simulations to verify, narrows down that list, and presents it as an answer for further research,” said DOE official Gutsche.
Fermilab’s reputation for handling massive data streams makes it a natural fit. The FDP is built on thousands of hard drives and already supports data from CERN’s Large Hadron Collider CMS experiment, the Short Baseline Neutrino program, quantum‑information research, microelectronics development, and advanced theory work. It is also preparing to host data for the upcoming Deep Underground Neutrino Experiment.
According to FDP’s lead partner, Chin Guok, “Data is the common denominator behind major scientific endeavors, and AI is fundamentally data‑driven. To train and run AI models, you need large volumes of data. Fermi Data Platform can support AI training and inference on large scientific datasets.” The platform’s architecture prioritizes repeatable, high‑throughput access that AI‑supported research demands.
The American Science Cloud will layer the FDP’s storage with AI services that let researchers describe a need and have AI tools tap into national‑lab resources—including supercomputers, scientific datasets, and simulation capabilities—without manual coordination. The cloud’s design is intended to be integrated, so a scientist can request a task and receive a refined set of results, such as a narrowed list of candidate battery materials.
Fermilab’s involvement was enabled by the laboratory’s recent management transition in 2025, which brought the Fermi Forward Discovery Group (FFDG) to oversee operations. The laboratory’s expertise in data‑management and its existing Tier‑1 role in the Worldwide LHC Computing Grid positioned it as a natural fit for Genesis’s data needs.
A key challenge for AI in science is that raw data from instruments often lacks the structure and metadata required for machine‑learning models. The FDP bridges that gap by storing datasets in an AI‑ready format and providing tools for data discovery and access.
Genesis’s broader strategy includes deploying AI on 26 national‑science and technology challenges with the full $37 billion budget. The American Science Cloud is one of the mission’s flagship platforms, intended to be a shared infrastructure that supports researchers across disciplines—from high‑energy physics to materials science to fusion‑energy research.
While the FDP’s exact storage capacity is described only as petabytes, the platform’s scale is sufficient to hold the large datasets generated by CMS, neutrino experiments, and quantum research. Its secure, large‑scale infrastructure is expected to accelerate AI‑driven discoveries by making data readily available for training and inference.
The partnership is still in its early stages. Fermilab has begun offering data‑storage and data‑access tools engineered for AI‑supported research, and the American Science Cloud is developing the AI services that will interface with the FDP. Next steps involve integrating the platform into the cloud’s workflow, testing AI pipelines on real scientific data, and expanding the data catalog to include additional experiments.
In summary, Fermilab’s Fermi Data Platform is a key component of the DOE’s Genesis Mission, providing the storage, tools, and expertise needed to make large scientific datasets AI‑ready. The partnership aims to speed up the scientific process by automating data discovery, simulation, and result filtering, thereby enabling researchers to focus on insights rather than logistics.
