The U.S. Department of Energy (DOE) is laying the groundwork for a public-private consortium designed to unify scientific data from its national laboratories to train “self-improving” AI models.

In a Request for Information (RFI), The DOE said these models must be made available to the scientific community through government, academic, and private-sector programs and infrastructure, including cloud technologies. 

“This historic mobilization of DOE, the National Laboratories, and private partners will serve as a force multiplier in executing America’s AI Action Plan to achieve global dominance in AI, and to advance scientific discovery, energy, and national security,” DOE said.

DOE is seeking feedback from AI developers interested in using its data or expertise; organizations of any size willing to partner with DOE and the National Laboratories to build advanced AI models; and contributions from think tanks, investors, research groups, and other interested stakeholders.

The RFI also asks stakeholders how DOE can effectively mobilize national labs to collaborate with industry, ensure scientific data is properly prepared for AI models, and structure a public-private consortium that supports diverse scientific disciplines. 

It seeks input on delivering AI models through cloud-based programs to accelerate scientific and energy innovation.