Mulholland: Scientific AI gains most value from unpublished negative experiment results
Greg Mulholland · How AI is rapidly advancing new materials for clean energy · Jan 23, 2024 · at 14:03
Citrine Informatics CEO Greg Mulholland explains why standard peer-reviewed scientific papers provide poor training data for AI models compared to raw industrial laboratory experiment histories.
0:00 / 0:26exact quote · 26.0s
720p mp4 · rendered on demand · StarZero watermark
“Unfortunately, from a scientific standpoint, the most valuable thing to AI is the two examples. But you know that those scientists did not just do two experiments. You know that those scientists did months and months and months of work and refinement and failed dozens of times to get a result that they thought they could, only to succeed later. And there's learning there in those what are known as negative results.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Greg Mulholland
Mulholland: Room-temperature superconductors will require new physics
“My prediction is that it will require at least a variant on known physics, if not entirely new physics, to make happen, just because that's sort of the nature of those massive breakthroughs, especially, I mean, room temperature superconductor is, A guaranteed …”
Mulholland: Energy transition will fail without accelerated materials innovation
“Candidly, I don't think we get there by a long shot. Even with new materials, it's a heavy lift because there are those cultural and cost and systemic issues we need to overcome.”
Mulholland: China pushes LFP batteries while US relies on high-risk cobalt chemistries
“In China, there's a big push for lithium iron phosphate. In the U.S., we have, you know, sort of standard lithium ion batteries that have a lot of nickel, manganese, and cobalt in the materials, and those things are hard to source, and rising in cost, and some…”
Mulholland: High lab experiment costs prevent large data volumes in materials
“One of the things that we don't have in the materials industry is huge volumes of data because everything costs money, right? You know, there is no free experiment.”