Google to commit $40 million in AI tokens and cloud credits to the Genesis Mission science program
Google has pledged $40 million in AI tokens and cloud credits to the Genesis Mission, a US Department of Energy science program launched by presidential executive order on 24 November 2025. Researchers at the national laboratories will get access to Gemini models and Google's compute resources to accelerate work in energy, quantum technologies and materials science.
AI-processed from DeepMind Blog; edited by Hamidun News
Google has announced $40 million in AI tokens and cloud credits for the Genesis Mission, a US government science initiative launched by presidential executive order on 24 November 2025 to accelerate discovery with artificial intelligence.
What is the Genesis Mission
The Genesis Mission is a US national program run by the Department of Energy (DOE). It combines artificial intelligence, supercomputers and federal laboratory data into a single research platform. The goal stated in the 24 November 2025 executive order is to multiply the productivity of American science in priority areas: fusion energy, quantum technologies, new materials and biotechnology.
- $40 million is the size of Google's commitment in AI tokens and cloud credits
- The Genesis Mission was launched by US presidential executive order on 24 November 2025
- The program is run by the US Department of Energy (DOE)
- The initiative involves the DOE national laboratories
- Researchers will get access to Gemini models and Google's compute resources
What Google's contribution provides
Google will provide the $40 million as resources rather than cash: tokens for using its AI models and credits for cloud computing. According to DeepMind, researchers working under the Genesis Mission will get access to the Gemini family of models, DeepMind's scientific tools and Google Cloud infrastructure.
The "tokens and credits" format means the funds are tied to Google's platform. Laboratories spend them on specific model calls and cloud resources instead of receiving a free-to-use budget. This takes the cost of expensive inference and compute off research groups, a line item that usually eats a noticeable share of scientific grants.
Why Big Tech is taking part
The Genesis Mission was built from the start as a public-private partnership: the government supplies laboratory data and supercomputers, while technology companies supply models and cloud. Google is not the only participant, but $40 million is one of the larger corporate contributions at the program's launch.
For Google, it is a way to establish Gemini and Google Cloud in the US government's scientific sphere, a segment previously dominated by specialized HPC contractors. As DeepMind's blog notes, the company sees science as one of the key areas for applying AI, from predicting protein structures to modeling physical processes.
"Accelerating the frontier of scientific discovery" is the title of
DeepMind's official blog post on Google's contribution to the Genesis Mission.
What this means
Major AI labs are embedding themselves ever more deeply in government science. Instead of one-off grants, Big Tech provides access to models and cloud in exchange for a presence in strategic research. Google's $40 million signals that competition for the US scientific AI market has moved from declarations to concrete resource commitments.
Frequently asked questions
What is the Genesis Mission?
The Genesis Mission is a US Department of Energy science program launched by presidential executive order on 24 November 2025. It combines AI, supercomputers and national laboratory data to accelerate research in energy, quantum technologies and materials science.
In what form is Google providing the $40 million?
Google is providing not cash but $40 million in AI tokens for using Gemini models and credits for Google Cloud computing. Researchers spend them on model calls and compute resources for their projects.
Access Frontier LLMs & Multimodal AI in One Workspace
Discover NomadNet AI (nomadnet.ai) — unified access to frontier language models, research synthesis, and multimodal workflows.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.