Your TL;DR: DOE just released a fast-moving, $293M funding opportunity for AI-driven R&D across science and energy domains. It is one of the few near-term paths to non-dilutive capital while SBIR timelines settle. The catch is structure: multi-institution teams, national lab partnerships, and cost share requirements will determine who is ready to compete and who is not.
A rare window for AI-driven R&D funding
There is a noticeable gap forming as teams wait for SBIR cycles to stabilize. The Department of Energy has stepped into that gap with a substantial funding opportunity focused on artificial intelligence applied to scientific discovery and energy innovation.
Learn more about the solication: https://science.osti.gov/-/media/grants/pdf/foas/2026/DE-FOA-0003612.pdf?utm_medium=email&utm_source=govdelivery
Learn more about the Genesis Mission: https://genesis.energy.gov/
The scale is meaningful. Roughly $293 million is available, with Phase I awards ranging from $500,000 to $750,000 and Phase II expected to scale at three to five times that amount. The structure moves quickly, with Phase I projects lasting nine months and potential Phase II efforts extending to three years.
The timeline alone should get attention. Applications are due April 28, with project starts anticipated July 1. That pace is unusual for federal funding and signals intent to move projects forward without delay.
If you are considering whether this fits your roadmap, it is worth taking a serious look at how your current R&D aligns with AI-enabled discovery and whether your team structure meets DOE expectations.
What DOE is actually looking to fund
This is not general AI funding. The focus is tightly tied to advancing scientific workflows and accelerating discovery.
Projects that tend to resonate include AI models that improve predictive accuracy, reduce experimentation time, or automate complex analysis. That can show up as digital twins, simulation environments, or integrated AI systems embedded within experimental or computational pipelines.
The topic areas are broad enough to capture a range of industries:
- Advanced manufacturing
- Biotechnology
- Critical materials
- Nuclear fission and fusion
- Quantum information science
- Semiconductors and microelectronics
- Energy systems and discovery science
What connects these areas is not the sector itself but the expectation that AI meaningfully changes how research and development are performed.
DOE reviewers will not be persuaded by general claims about AI potential. They look for measurable improvements. Faster processing, stronger predictive power, or clear automation gains tend to separate competitive applications from the rest.
The structure is where most applications break down
The technical idea is only one part of the equation. The structure of the team often determines whether a proposal is even viable.
This opportunity requires multi-institutional collaboration. Phase I teams must include at least two of the following:
- A DOE or NNSA national lab or user facility
- Industry
- An academic institution, nonprofit, or similar entity
Teaming with a national lab is not optional in practice. It is central to how DOE evaluates feasibility and alignment with its infrastructure.
There is also a financial consideration that tends to surface late for many applicants. Commercial entities should plan for a 20 percent cost share. That requirement can quietly shift a project from feasible to constrained if it is not accounted for early.
This is often where strong concepts stall. The science is compelling, but the team is incomplete or the financial structure is not fully thought through.
The GAP that most teams underestimate
Many teams assume they can assemble partnerships after identifying the opportunity. In practice, the strongest proposals come from teams that have already worked together or have clearly defined roles and integration points before writing begins.
Waiting to form those relationships introduces friction. Alignment across institutions takes time. Data access, intellectual property considerations, and workflow integration need to be understood early, not negotiated under deadline pressure. That gap between concept and coordinated execution is where otherwise competitive ideas lose momentum.
What makes an application competitive
DOE tends to reward clarity and alignment over ambition alone.
Strong applications usually demonstrate:
- A clear advantage gained through AI, not just its inclusion
- Quantifiable improvements tied to the proposed approach
- Use of DOE data, infrastructure, or national lab capabilities
- A team that reflects real interdisciplinary integration
- Alignment with defined topic areas and mission priorities
Phase I, in particular, leans toward proof-of-concept work that can show a credible path to scale. Reviewers want to see early indicators that the approach can extend into a greater, more impactful Phase II effort.
Additional benefits that are easy to overlook
Funding is only part of the value. Awardees may gain access to national laboratory resources, datasets, and specialized infrastructure. There is also potential integration into initiatives like the American Science Cloud and participation in broader collaborative ecosystems. These elements can shape long-term R&D capacity in ways that extend well beyond the initial award.
Timing and next steps
The deadlines are fixed and close:
- Phase I applications due April 28, 2026
- Phase II letters of intent due April 28, 2026
- Phase II applications due May 19, 2026
- Phase II follow-on from Phase I due December 17, 2026
Selection timing is not specified, though the July 1 project start suggests an expedited review process.
If you are evaluating this opportunity, now is the time to pressure test your concept against DOE’s expectations and your team structure against the program requirements.
If you are weighing whether this opportunity aligns with your current R&D trajectory, it is worth mapping your project against DOE’s evaluation criteria and identifying where gaps may exist.
Where this fits while SBIR is in flux
This opportunity is not a replacement for SBIR. It operates differently, both in structure and expectations. That said, it offers a near-term path to non-dilutive funding at a moment when many teams are waiting for clarity elsewhere. Organizations that can move quickly, form the right partnerships, and clearly articulate how AI changes their R&D approach are positioned to take advantage of this window.
Teams that delay or treat this as a standard grant application often find themselves outpaced by groups that understand how DOE evaluates both science and execution. As you consider whether to pursue this, think less about eligibility and more about readiness. The distinction tends to determine outcomes.
If you are preparing to pursue DOE funding, consider how your current partnerships, data strategy, and technical approach will hold up under the level of scrutiny this program requires.
A note on how EBHC approaches this work
EBHC operates on a flat, fee-for-service model. There are no success fees, contingency structures, or percentage-based compensation tied to federal awards. The scope of work drives the pricing, based on the level and type of support required.
Intellectual property stewardship is treated with the same level of care as the proposal itself. The team is entirely U.S.-based, structured as W-2 employees, and operates under organization-wide NDA protections when engaged. Ownership remains clear and unchanged. Your innovation stays yours, the grant stays yours, and all associated data and rights remain with your organization.
This approach tends to matter more in opportunities like this one, where collaboration is required but control still needs to be well defined.
Ready To Take the Next Step?
We assist our clients in locating, applying for, and evaluating the outcomes of non-dilutive grant funding. We believe non-dilutive funding is a crucial tool for mitigating investment risks, and we are dedicated to guiding our clients through the entire process—from identifying the most suitable opportunities to submitting and managing grant applications.