Conversations from the Ground: Dr. Thomas Danielson

Dispatch 003: Artificial Intelligence Operational Usage
Date: May-Present
Location: Washington, D.C.
Associated with: The Gigawatt Project
Type: Field Transcript
Status: Published

Savannah River National Lab | Special Competitive Studies Project AI Expo | Washington D.C. May 2026

PT 1

In May, I had the honor to sit down with Savannah River National Lab scientist Tom Danielson. Finding myself in Washington, D.C. for the Special Competitive Studies Project AI Expo, I sought to learn how Artificial Intelligence was being implemented on the national level. With the talk of AI in the mainstream regarding consumer facing AI and LLMs being the majority of conversation, I was inclined to discover more about what was happening at the larger scale.

In dispatch 003 and the first of The Gigawatt Project, I present here a partial transcript from my radio broadcast and will release the full conversation over a series of dispatches.

Source: Personal recorded interview with Tom in May 2026. Which was edited for special on-air broadcast use for Uncontaminated Sound WGXC 90.7 FM June 2026. Machine transcript (Adobe Premiere Pro), lightly cleaned of filler; content unedited.The broadcast version is linked at the bottom of this page.

Tom Danielson: Our our motto is put science to work. And so, what we're looking for are those technologies that can be packaged and put into the field to realize some return on investment. Right away. Of course, we do research across the whole technology readiness level spectrum. My particular focus area is in, a lot of the work that goes on within the Office of Environmental Management.

So this is dealing with, the legacy, nuclear waste. You know, dating back to the Manhattan Project. Four eras. And then, of course, going on here today. So this is a long mission duration. There's, a number of different mission areas there in, one of the big areas of emphasis right now is, liquid waste processing.

So there are 90 million gallons of, highly heterogeneous liquid radioactive waste that's present in underground waste tanks, at both the Savannah River site as well as the Hanford site, that need to be processed. So they've run it through this very big, complex, flow sheet of, of chemical processing operations, ultimately to try to get it into a stable waste form that's either a glass or a grout.

Rob Lundberg: You just heard Tom Danielson from the Savannah River National Laboratory. I'm Rob Lundberg, host of Uncontaminated Sound on WGXC 90.7 FM. This is a broadcast from the field at the Special Competitive Studies Project AI Expo, which occurred early May in Washington, D.C. I went to the expo to understand how decisions about artificial intelligence are being made at the national level, and what that means for infrastructure and the systems we depend on.

Rob: And now, how has the technology helped in making the process cleanup process more efficient? Has it all. Or has it just been more data collection and more, decision oriented? Have you seen efficiencies in the processes of, cleaning up?

Tom: So I think I think today it's still very, very early on. Certainly there are some success stories, across these different contractors in terms of how they've deployed AI and how they've, decreased the duration of, of decisions that they will ultimately make. By bringing all this data together. So you hear a lot of stories at this point about, you know, bringing data sets together, to, to make the sort of the chemical engineering decisions that used to take months. Coming down and being compressed into days, at this stage.

Rob: So really making the processes more efficient.

Tom: Exactly.

On Genesis Mission

Rob: And now, can you touch upon the Genesis, mission a bit? And how's that affected your mission now?

Tom: Yeah, absolutely. So, the Genesis mission, of course, is is kind of a paradigm shift almost for the Department of Energy, right? A large scale platform, that comes with full hardware, as well as software infrastructure that can be tremendously beneficial across a lot of Doe mission areas. From the environmental management perspective, we see a really great, sort of feedback loop there. So, there's lots of data that's been a lot of different areas. So the subsurface, you know, groundwater, flow and transport, being able to make really good models there that can then extend out to, the energy sector. Gas as well as critical minerals. And then when we come back up to, to the facilities, kind of these processing facilities. Again, the types of, technologies that we're developing can be deployed in any chemical processing facility across the country.

We just have one particular domain we're looking at. So where the opportunity comes is that we for the Department of Energy, Office of Environmental Management, we can develop these technologies. But then the Genesis mission provides that opportunity to take those and generalize them and make that resource available to the broader scientific and engineering community. So I think there's a lot of opportunity there.

And then the data sets. Training new models. Foundational models for subsurface, for operational facilities. For so many different areas. There's so, there's so much, so much there. In addition to that, there's so much data, right. And so a lot of what Genesis is doing right now is creating these generalizable techniques.