Big Ideas Lab podcast examines how AI Is reshaping science at LLNL

As the world moves from the Information Age to the Data Age, AI is not only revolutionizing society — it’s fundamentally transforming how researchers at Lawrence Livermore National Laboratory (LLNL) approach science.

On the latest episode of The Big Ideas Lab podcast, listeners will get an inside look at this transformation, as LLNL physicists Brian Spears and Kelli Humbird explain how AI is no longer just about chatbots and automating simple tasks. At the Lab, AI is accelerating scientific discovery, combining physics, data and machine learning to model complex, real-world phenomena like inertial confinement fusion (ICF), drug design and advanced manufacturing with unprecedented speed and accuracy. Listen on Apple or Spotify.

A key approach to AI at LLNL is “cognitive simulation,” or CogSim, which combines high-performance computing, experimental data and AI models to generate and test scientific ideas faster than ever before. This approach assisted in one of LLNL’s finest achievements: fusion ignition. By processing enormous amounts of simulation data, AI helped predict which ICF target and experimental designs were most likely to succeed.

“The CogSim piece said, ‘looking at all of the simulations and the data we have from the past, I've got a capability to analyze new designs,’” Spears recalls of that fateful moment in Dec. 2022. “We showed those tools a new design, and it said, ‘you've got a greater than 50% chance of igniting’ … It’s not overwhelming confidence, but for the prior six decades, that number had been tiny.”

But predicting and advancing fusion energy with AI is just the beginning. The episode dives into how AI is also transforming drug discovery — cutting years off the timeline to identify promising new treatments. In collaboration with biotech and pharmaceutical companies, along with other government entities, LLNL’s AI-powered platforms have already led to the development of new drug candidates for cancer and inroads against evolving viruses like COVID-19.

LLNL’s AI efforts are also being strengthened by public-private partnerships with companies like OpenAI, Anthropic and others. These collaborations bring cutting-edge AI tools like reasoning models into the Lab’s scientific workflows, helping researchers generate hypotheses, simulate outcomes and zero in on the most promising experiments and designs across a wide range of fields.

Still, as Humbird emphasizes, these new tools are there to augment — not replace — human judgment or scientific rigor.

“These are tools that can help you do your job more efficiently or faster, but they're not replacing the final human analysis of the decisions we're about to make or of the experiment we're about to field,” Humbird says.

As the episode reveals, LLNL’s integration of AI with science is opening doors to faster, smarter and more scalable experimentation. Whether it’s fusion energy, medicine, advanced manufacturing or materials science, Lab researchers are showing what’s possible when human creativity meets the power of AI.

Listen to the latest Big Ideas Lab podcast here on Apple or Spotify.