Keith Ulmer

Keith Ulmer writes for CERN: Nobody knows what new physics will look like

Kirsten Apodaca — Wed, 30 Sep 2026 16:17:02 +0000

Keith Ulmer writes for CERN: Nobody knows what new physics will look like Kirsten Apodaca Wed, 09/30/2026 - 10:17

Illustration by Sandbox Studio, Chicago

Keith Ulmer

Opinion by Keith Ulmer, an Associate Professor of Physics at the �Ǹ�Թ�. Republished from Beyond Standard with permission from CERN.

Every year, theoretical physicists dream up hundreds of ideas for how new physics could materialize in the data produced by the Large Hadron Collider. As a physicist on one of the LHC’s biggest experiments, I love the enthusiasm. But testing each one of these ideas takes time—often, the length of an entire PhD thesis. We just don’t have the people power to hack away at all these clever ideas individually. Even more, the data volume generated by the LHC’s collisions is extraordinary; far beyond what we can record and store. We only have microseconds to select the most promising collisions, and if a theorist’s cool new idea doesn’t make the cut, the events that could prove it right are automatically dumped in the trash.

This has always made me and my colleagues on the CMS Experiment at CERN deeply uncomfortable. What if we’re missing out on something extraordinary simply because a new idea didn’t make our “this could be interesting” list?

Illustration by Sandbox Studio, Chicago with Thumy Phan

One idea is that—in addition to our “this could be interesting” list—we also keep a random selection of events that might not seem interesting at first glance, but could later reveal unexpected patterns. But new physics is so rare that catching it in a random sample is like throwing a fishing net into the bay and catching a kraken. Another solution could be to reduce the file sizes of the data we store, like going from high-resolution photographs to pixelated JPEGs. This would let us store a much higher fraction of events. But it would also mean that if we do find evidence of new physics, we would only have a blurry image of it.

So what can we do?

Thanks to a Genesis Mission award from the US Department of Energy, which was granted to my group at the �Ǹ�Թ�, and our collaborators at Fermilab, UCSD, and Johns Hopkins, we no longer have to compromise. We can have our selection of high-quality data. We can have our pixelated thumbnails. We can adapt the concept of “random sampling” so that it is no longer random, but optimized to search for the strangest, most amazing events. And we can do all of this while staying within our computing power constraints.

How? By developing ultra-fast artificial intelligence and deploying it at the earliest stages of data collection.

First, we are reimagining something we call an anomaly detection trigger. Triggers are hardware and software tools that automatically sort our data into “this is interesting and should be saved” or “this is boring and can be chucked.”

Traditional trigger systems search for pre-programmed patterns in the data. But for the anomaly detection trigger, we don’t tell it what to look for. We simply ask, “Does this event look different from all the others?” If the answer is yes, we flag and save it. This anomaly trigger will not replace our traditional triggers, but instead add a special “anomalous” data set that preserves the strangest, most amazing events. With this data set, we no longer need to individually test every new physics model; we can simply see if any weird patterns emerge.

Illustration by Sandbox Studio, Chicago

We deployed an initial prototype during LHC Run 3 and proved that the concept works. With the Genesis Mission grant, we can transform this idea from a proof of principle into full production and deploy it during the High Luminosity run of the LHC, which will create some of the most complex data ever seen and at a rate previously unimaginable.

Second, we are pushing a concept called data scouting. In addition to keeping our high-resolution data of the things we want to study (i.e., Higgs bosons), we also want to keep everything else—but in a much more compact, “pixelated” form. The concept is like covering a nature reserve with cheap, low‑resolution wildlife cameras. Sure, we will still have our high-resolution cameras set-up in places where we know we will see something cool, but we will also have the opportunity to catch a blurry tentacle unexpectedly emerging from the bay.

And this brings us to the final part: If we find an anomaly, we don’t just flag it. We use AI to reverse‑engineer what made those events strange and then write new trigger criteria so that we can capture that blurry tentacle in full resolution if it ever appears again. (And thus, maybe finally catch our kraken.)

This three‑part strategy—anomaly detection, data scouting, and AI‑assisted trigger redesign—will give us the flexibility to catch not only the strange things dreamed-up by theorists, but the events nobody saw coming. In this way, these tools are not just new technology: They are a paradigm shift. And for me, that’s the most exciting part of this research. We no longer have to pretend that we know what new physics will look like. We can openly embrace our own ignorance and search for physics signatures that lie beyond the human imagination.

In an opinion piece for CERN, Associate Professor Keith Ulmer explores how artificial intelligence could help physicists uncover previously unseen phenomena in particle collision data.

Off

Traditional

Illustration by Sandbox Studio, Chicago

White

Physics faculty awarded Department of Energy’s Genesis Mission funding for AI-based projects

Kirsten Apodaca — Tue, 18 Aug 2026 15:02:50 +0000

Physics faculty awarded Department of Energy’s Genesis Mission funding for AI-based projects Kirsten Apodaca Tue, 08/18/2026 - 09:02

Kirsten Apodaca

Six physics faculty from CU �Ǹ�Թ� are leading or contributing to new projects awarded more than $1.6 million in highly competitive U.S. Department of Energy Genesis Mission grants, the agency announced on July 22. Nationwide, only 278 phase-one projects were funded out of over 5,000 proposals.

Yuan Shi and Keith Ulmer are principal investigators on their respective projects, Dennis Perepelitsa is a co-principal investigator on two projects, Jamie Nagle is a co-principal investigator on another, and Xun Gao and Scott Parker are collaborators on the project led by Shi.

The Genesis Mission is a historic national initiative led by the U.S. Department of Energy, which is building the world’s most powerful integrated science discovery platform. By uniting government, industry, academia, and philanthropy, it is accelerating breakthroughs in energy, scientific discovery, and national security through a new platform that combines AI, supercomputing, quantum systems, and advanced scientific instruments.

“Having six CU Physics faculty involved across four Genesis Mission projects is incredibly impressive,” says Tobin Munsat, professor and chair of physics. “This reflects the broad strength of our department and demonstrates how our faculty are leading the way in putting AI and quantum technologies to work on some of the most challenging questions in fundamental science.”

AI for analyzing particle collisions at the Large Hadron Collider

Ulmer, with collaborators from the University of California San Diego, Fermi National Accelerator Laboratory and Johns Hopkins University, will use AI to analyze largely untapped datasets of particle collisions from the Large Hadron Collider (LHC) at CERN.

Each day, the LHC produces about 4,000 petabytes, or 4 billion gigabytes, of particle collision data used by physicists around the world to better understand the fundamental nature of the universe. Because of the enormous amount of data, current analyses are limited to roughly one in every 10,000 particle collisions, leaving potential anomalies undetected.

By combining high-energy physics expertise, data science, and industry AI methods, Ulmer and his collaborators will develop a system capable of analyzing the full dataset from the LHC. Called the AI-Inclusive Discovery of Anomalies in Scouting (AIDA-Scout), the system will analyze data in real time, automatically adjust to changing detector conditions, and find anomalies in collision data that could lead to new scientific discoveries.

Using quantum computing technologies and AI to simulate plasma for fusion pilot plants

Shi is leading a project to generate plasma simulations needed for fusion pilot plants. Collaborators on the project include Gao and Parker, as well as colleagues from Lawrence Livermore National Laboratory and Infleqtion.

Even today’s most powerful supercomputers cannot produce certain plasma simulations essential for developing fusion pilot plants.

The team will develop new computational methods combining quantum computing, quantum machine learning, and AI to create more efficient and accurate simulations needed for fusion to become a potential energy source.

“This project seeks to prepare the fusion community for the coming generation of quantum computers by creating algorithms and software that can take advantage of quantum hardware as it matures,” says Shi. “The resulting capabilities could significantly accelerate plasma simulations, supporting advances in fusion energy science and scientific computing.”

AI for understanding quark-gluon plasma

Some of the final collisions captured by the sPHENIX detector at the Relativistic Heavy Ion Collider, a nuclear physics research facility at Brookhaven National Laboratory. (Image Credit: Brookhaven National Laboratory)

A project led by Baruch College, City University of New York involves CU �Ǹ�Թ� physics faculty Jamie Nagle and Dennis Perepelitsa as co-principal investigators. The project’s leadership team includes Yeonju Go, a former postdoctoral researcher in CU �Ǹ�Թ�’s experimental nuclear physics group.

The project is designed to better understand the quark-gluon plasma, a phase of matter which only existed microseconds after the Big Bang at temperatures exceeding two trillion Kelvin. At Brookhaven National Laboratory, scientists recreate tiny droplets of quark-gluon plasma by colliding gold nuclei together at nearly the speed of light. These collisions are analyzed by the sPHENIX detector, which acts as a giant camera, capturing approximately 15,000 particle collisions each second.

When creating the quark-gluon plasma, sometimes there are two high energy quarks that scatter, and this project aims to separate and analyze these specific images by using an unsupervised AI framework known as cycle-consistent generative learning.

“We know how quarks scatter at high energy, but we want to understand how they scatter when they’re inside the plasma,” says Nagle.

AI for enhancing the foundation model for nuclear and particle physics

Dennis Perepelitsa is a co-principal investigator on a project led by colleagues at Lawrence Livermore National Laboratory aimed at enhancing the Foundation Model for Nuclear and Particle Physics (FM4NPP), a large-scale AI model built to extract scientific insights directly from raw detector data.

Foundation models go beyond the more commonly known large language models (LLMs) by being able to perform a variety of tasks. The project will extend the current FM4NPP, which only includes information from one sPHENIX detector subsystem, to include a variety of other detectors, making the model truly multi-modal and enhancing the capability of the detector for quark-gluon plasma measurements. Ultimately, the FM4NPP effort is envisioned as a multi-institution consortium between national labs, universities, and industry to accelerate scientific discovery in a broad set of nuclear and particle physics datasets.

“In modern collider detectors, physics processes leave signatures in multiple subsystems at once, and a comprehensive picture is needed for the highest scientific sensitivity. The foundation model is based on similar principles as commercial LLMs, which is that with enough data and scale, it can make connections beyond the existing non-AI approaches,” says Perepelitsa.

View the CU �Ǹ�Թ� press release

Off

Traditional

White