The Department of Energy is the US cabinet department responsible for national energy policy, nuclear security, and the network of 17 national laboratories. Its AI relevance runs in three directions: it is the lead agency of the Genesis Mission, the federal AI-for-science initiative; it is one of the largest operators of high-performance computing in the world through its national-lab supercomputers; and it is a central analytic authority on the electricity demand that AI data centers are placing on the US grid. Its national labs — including Lawrence Berkeley National Laboratory — produce some of the most-cited estimates of data-center power consumption.
Genesis Mission
DOE leads the Genesis Mission, the national AI-for-science initiative established by executive order in November 2025. The department describes the program as an integrated platform connecting its supercomputers, AI systems, experimental facilities, and scientific datasets across all 17 national laboratories, with partners in industry and academia, and states a goal of doubling the productivity and impact of US research and development within a decade (Source: energy.gov).
Implementation has proceeded through named workstreams and partnerships. The department announced 26 Genesis Mission "Science and Technology Challenges" spanning its mission areas (Source: energy.gov) and a cohort of roughly two dozen AI research partners (Source: fedscoop.com). Near-term AI capacity is being expanded through the Lux AI cluster at Oak Ridge National Laboratory, deployed in 2026 in partnership with AMD (Source: amd.com). In July 2026 the White House announced more than $5 billion in federal commitments to expand the mission (Source: whitehouse.gov), and the OSTP report Science: A New Golden Age recommended fully funding and scaling it as the flagship national AI-for-science initiative. Contributions from other actors include Microsoft's $60 million commitment ($40 million in Azure compute credits, $20 million in engineering) announced July 22, 2026, with initial projects at Pacific Northwest National Laboratory, Lawrence Livermore National Laboratory, and Johns Hopkins (see Microsoft), and the Defense Department's planned commitment of more than $150 million across fiscal 2026–27 for a Genesis-aligned high-performance computing environment (see DOD — Department of Defense (AI Deployer)).
The White House announced the mission's first awards on July 23, 2026, directing more than $5 billion across 278 projects in all 50 states and 342 participating institutions, including 16 DOE and NNSA national laboratories (Source: theverge.com). DOE Under Secretary for Science Dario Gil said by July 24, 2026 that the department will issue additional purpose-driven "challenges" under the mission at an upcoming stakeholder summit (Source: insideaipolicy.com).
Compute and national laboratories
Through its national laboratory system, DOE operates some of the world's largest supercomputers and serves as the federal government's principal scientific-computing deployer. Argonne and Oak Ridge national laboratories lead AI research projects under the Genesis Mission (Source: anl.gov), and the department frames the combined platform of supercomputers, AI systems, and scientific instruments as a single integrated research instrument once complete (Source: energy.gov).
Proposed restrictions on foreign-national lab access
The GATE Act, introduced August 14, 2026 by Representative Pat Harrigan, would bar foreign nationals from China, Russia, Iran, North Korea and Cuba from accessing national laboratory facilities, information or technology, subject to Energy Department waivers (Source: nextgov.com). The bill would place a personnel-access control on the same laboratory system that hosts the Genesis Mission computing described above, and its waiver provision leaves the operative scope to departmental discretion. See AI and National Security and US-China AI Competition: Different Races, Different Metrics.
Data-center electricity demand
In analysis reported June 24, 2026, the DOE's Lawrence Berkeley National Laboratory projected that data-center electricity demand could exceed 15 percent of US consumption by the end of the decade, a figure the department used to underscore its call for efficiency gains rather than supply expansion alone (Source: insideaipolicy.com). The projection sits alongside the Federal Energy Regulatory Commission's parallel June 24 move pressing grid operators to revise market rules for large power users, the two actions framing the federal energy-and-AI question around both how much power data centers will need and who pays for the grid to serve it (see Data Center Siting / AI Power Politics).