Google DeepMind supports U.S. Department of Energy on Genesis: a national mission to accelerate innovation and scientific discovery

Source: DeepMind Date: 2025-11-24 URL: https://deepmind.google/blog/google-deepmind-supports-us-department-of-energy-on-genesis/

Summary

Google DeepMind joined the White House’s Genesis Mission providing frontier AI tools to DOE’s 17 National Laboratories: AI Co-scientist (multi-agent hypothesis generation on Gemini), AlphaEvolve, AlphaGenome, WeatherNext, and Gemini 3 for Government. Concrete results: AI Co-scientist proposed liver fibrosis drug repurposing candidates validated experimentally and predicted antimicrobial resistance mechanisms matching lab results before publication — compressing hypothesis development from years to days. Historical tie: DOE’s Brookhaven National Laboratory work was foundational to AlphaFold (2024 Nobel Prize in Chemistry).

Implications

The DOE partnership is the most significant US government AI alignment to date. 17 National Laboratories covering nuclear, energy, climate, and biomedical research with direct access to Gemini 3 + AlphaEvolve + AlphaGenome is an unprecedented government-AI lab integration. It positions Google as the primary AI infrastructure provider for US national science.

AI Co-scientist’s liver fibrosis result is the most concrete claim. Proposing drug repurposing candidates that survived experimental validation is a real scientific outcome, not a benchmark. If reproduced across other disease areas, it’s the beginning of AI-accelerated drug discovery at government scale — which has major implications for NIH and pharmaceutical R&D timelines.

Nobel Prize citation is a calibration move. Noting that DOE’s Brookhaven work was foundational to AlphaFold’s Nobel Prize is Google saying: we’ve already done Nobel-caliber science with government institutions. Genesis is the next chapter of that same partnership.

Watch:

• Whether Genesis Mission produces peer-reviewed publications citing AI Co-scientist contributions
• Microsoft/OpenAI and Anthropic competing bids for national laboratory AI partnerships
• Policy implications: if AI accelerates government science, what’s the acquisition and IP structure for discoveries made with commercial AI tools?