Google commits $40 million in AI credits to the DOE's Genesis Mission

Google DeepMind and Google Cloud committed $40 million in AI tokens and cloud credits to the U.S. Department of Energy's Genesis Mission, deepening a partnership that already gives DOE's national laboratories early access to Google's frontier AI tools.

What's new

Speaking at the DOE Genesis Mission Summit 2026, Google said: "Today, at the DOE Genesis Mission Summit 2026, we are expanding this by committing $40 million of AI tokens and cloud credits for researchers in support of the Genesis Mission." The commitment was announced jointly by Pushmeet Kohli, Google DeepMind's VP for Science and Strategic Initiatives and Chief Scientist for Google Cloud, and Karthik Narain, Google Cloud's Chief Product and Business Officer.

Alongside the funding, Google is extending Gemini for Government access across the DOE's National Laboratories: "We will provide Gemini for Government seats and tokens for one year to tens of thousands of users across the DOE National Laboratories' operations, research, and management teams." Google DeepMind also confirmed an early access program giving AI-for-science tools to all 17 DOE National Laboratories, while Google Public Sector positioned Gemini for Government as a potential AI backbone for the department's broader operations.

Context

The Genesis Mission is the federal government's initiative to mobilize DOE's 17 national laboratories, along with industry and academic partners, into an integrated AI-driven research platform, targeting problems like fusion plasma dynamics, new materials discovery, and processing the exabytes of data generated by the country's experimental science facilities. Google DeepMind first confirmed its involvement with the effort in a blog post last November. Meta separately detailed its own contribution to the initiative days earlier, offering its open-source vision models to a DOE scientific-imaging pipeline. Google's $40 million commitment and government-wide Gemini rollout mark a significant escalation from that initial research-collaboration posture into a funded, laboratory-wide infrastructure deal.

Why it matters

The commitment shows Google converting a policy relationship with the DOE into direct compute and subscription spend at national-lab scale, positioning Gemini as core infrastructure across federal science operations rather than a pilot tool for individual research teams. Providing free tokens and seats to "tens of thousands" of lab personnel is a much larger distribution push than the earlier early-access program covering just the 17 labs' research arms, and it sets up direct competition with Meta and other AI labs vying to become the default AI layer inside U.S. federal science infrastructure. For DOE researchers working on fusion, materials science, and large-scale experimental data, the deal removes a major cost barrier to using frontier AI tools in day-to-day research.