Physicists built the Large Hadron Collider to study the inner workings of the universe. Inside a 27-kilometer underground ring straddling the French-Swiss border, the machine smashes protons together at nearly the speed of light to produce—fleetingly—the smallest constituent building blocks of nature. Sifting through snapshots of these collisions, LHC researchers look for new particles and scrutinize known ones, including their most famous find, in 2012: the Higgs boson, whose behavior explains why other fundamental particles like electrons and quarks have mass.
Less well known is the intricate software engine that powers such discoveries. With particle collisions occurring at approximately a billion times per second, the facility generates about 40 terabytes of data per second, according to LHC physicist Maurizio Pierini. Consequently, researchers have developed extensive tools for extricating signal from noise. Part of the analysis involves creating detailed simulations of the collisions, which allow physicists to lay out exactly what they know, to more easily spot the unexpected. These simulations require massive computing power, and researchers will accelerate their data production as they prepare to upgrade the LHC to produce around six times as much in 2027. To scale up the required simulations, Pierini is considering the tech trend of the moment: artificial intelligence.
Specifically, Pierini has begun simulating proton collisions with an AI algorithm known as a generative adversarial network, or GAN. These algorithms are known for their ability to create realistic-looking fake data. “We can even simulate the behavior of particles that we don’t know exist,” says Pierini.
In fact, perhaps the most notorious application of GANs also involves producing objects that don’t exist. They are the software behind deepfakes, the fake images of human faces which have been used to forge videos that portray events that never really happened, or to glue a person’s image onto another person’s body. That’s right; the same algorithms that trolls use to make fake celebrity porn may also help uncover the universe’s deepest mysteries. Physicists have found a noble purpose for some of AI’s creepiest tools.
And this poses a moral quandary to physicists, who know that any algorithm can be both creepy and noble, depending on its context. “I’m quite convinced that AI will ultimately be either the best thing or the worst thing ever to happen to humanity, depending on how we use it,” says physicist Max Tegmark of the Massachusetts Institute of Technology.
Here’s another example of the amorality of AI algorithms: When applied to personal data, they can make racially biased mistakes. Studies have shown that commercial facial-recognition software makes mistakes with Black faces at far higher rates than with white ones. In January, a Michigan police department facial-recognition algorithm misidentified Robert Julian-Borchak Williams, a Black man, for shoplifting, resulting in possibly the first example of a wrongful arrest in the US due to faulty facial-recognition software. But physicist Brian Nord of Fermilab uses algorithms derived from facial-recognition technology to identify abnormal-looking galaxies whose light is warped by curved spacetime. Nord, who is Black, studies these galaxies for clues to why the universe is expanding faster.
In 2014, before Nord began using AI, he and his colleagues classified these galaxies by eye. “It was a huge endeavor to identify these objects,” says Nord. “So I started exploring other algorithms out there and accidentally came upon facial recognition.” The software worked—after all, to a computer, faces and galaxies are both just data. Yet for Nord, this raises concerns about how such work might escape from the lab: What if physicists made an algorithm for galaxies, but somebody else re-purposed it for surveillance?
In AI’s current state, Tegmark sees potential parallels with a shameful physics legacy: his field’s contributions to the development of nuclear weapons. In the 1940’s, physicists applied their fundamental discoveries about atomic nuclei to enable the bombings at Hiroshima and Nagasaki. Their technology culminated in existential and international tensions that continue today. “I feel quite responsible that my tribe, physicists, have contributed to the fact that we’ve almost had an accidental nuclear war now between the US and Russia over a dozen times,” says Tegmark.
With this historical perspective, he thinks physicists have a unique role to play in ethical AI development. Some physicists, including Tegmark and Nord, have taken up the mantle by fixating on one main problem: figuring out why these algorithms even work. Experts frequently use the term “black box” to describe AI: The computer can differentiate a cat from a dog, but it’s not clear how it got to the right answer. “I believe if we make our AI systems more intelligible, then we’ll be able to trust that they’re going to do what we actually want them to do,” says Tegmark.
For example, say you use an AI algorithm to “train” a computer to construct a mathematical model, or set of equations, that describes a cat, by showing it thousands of cat photos. The computer “learns” this model through a complicated process involving the repeated adjustments of sometimes millions of variables. Then, by calculating how well a new photo matches this model, the computer can determine whether it contains a cat. But if you have the computer construct this cat model multiple times, training it with the same set of cat photos, it produces a slightly different model each time. The computer might be comparably competent at identifying cats with each model, but be identifying them using subtly different mathematical arguments.
It’s unclear why the inconsistencies arise, or even how to talk about them. “My friends and I will talk about how machine learning research sometimes seems a little unscientific because a lot of the results aren’t reproducible or super quantified,” says physicist Savannah Thais of Princeton University, who has used AI algorithms to analyze LHC experimental data.
Thais recently published an essay for the American Physical Society encouraging physicists to participate more actively in AI ethics conversations. In the essay, she writes that physicists could play a critical role in improving algorithm “interpretability”—to carefully analyze algorithms and their inconsistencies to ultimately make them fairer. (Disclosure: this reporter writes for the American Physical Society.) In other words, Thais thinks that physicists can unpack how the AI sausage is made. “That’s such a big focus in physics work already—how precisely certain we are that something works the way we think it does,” she says. Physicists are already compelled to break down any object into its most fundamental pieces and interactions; Thais wants them to direct that compulsion to AI algorithms.
One strategy for unpacking AI involves training algorithms with data that physicists understand in excruciating mathematical detail. For example, Nord and a colleague, João Caldeira, studied how a neural network processes simulated data of a pendulum swinging back and forth. Physicists have understood pendulum behavior for centuries. Because the data is so simple, Nord and Caldeira can more easily track, step-by-step, how the algorithm organizes and processes the data.
Nord and Caldeira essentially trained their algorithm to estimate the strength of Earth’s gravity from their pendulum’s motion. But they included human fallibility in their training data: imprecise measurements that might correspond to a flawed measuring tape used to measure the pendulum’s length, or the sweaty high school student whose fingers slip while making the measurement. They then studied how the algorithm processed these errors into its final estimate of gravity.
While their understanding of algorithm error is still in early stages, their ultimate goal is to help develop a clear and accurate way to communicate an algorithm’s margin of error, a new area of AI research called “uncertainty quantification.” “It’s this idea of rigorously understanding AI’s error bars,” says Nord.
Tegmark is tackling AI inscrutability with a different strategy. Instead of deciphering an algorithm’s process in granular detail, he focuses on re-packaging its complicated output.
In a way, Tegmark treats AI algorithms like we treat human intuition. For example, a child learns to catch a ball intuitively, without ever needing math. But in school, she might learn to describe her intuition using equations of parabolas. Tegmark thinks that scientists should think of AI as providing intuition, like the instinctive ability to catch a ball, that should then be re-packaged into elegant, comprehensible math equations. Among computer scientists, this re-packaging is called “symbolic regression.”
For example, Tegmark and an MIT graduate student, Silviu-Marian Udrescu, fed data about planetary orbits to an AI algorithm. Conventionally, an AI algorithm would identify patterns in that data and represent them with some long, murky formula for the trajectory of a planet around a star. Tegmark’s algorithm, however, took the extra step of translating that esoteric formula into Kepler’s third law of planetary motion, a concise equation that describes a planet’s orbit in relation to its mass and a few other variables. Publishing in Science Advances this April, they call their algorithm “AI Feynman,” because it successfully re-discovered 100 equations from physicist Richard Feynman’s classic introductory physics lecture series.
By making AI processes more transparent, physicists offer a technical solution to a particular ethics challenge. But many ethical challenges cannot be solved with technical advances, says AI ethicist Vivek Nallur of University College Dublin. Ethical dilemmas, by nature, are subjective, and often require people with conflicting priorities to settle their differences. These people may disagree with an algorithm’s recommendation simply based on cultural or personal preference. “For a problem to qualify as a problem for AI ethics would require that we do not readily know what the right thing to do is,” writes philosopher Vincent Müller in the Stanford Encyclopedia of Philosophy.
For example, a 2018 MIT study of people in 233 countries found that participants’ reactions to ethically grey situations were culturally dependent. The study, presented as a game, asked participants variations of the trolley problem: In one case, should a self-driving car swerve to save its three passengers, including a child, and kill four elderly pedestrians? The researchers found that participants from cultures that emphasize the collective, such as in China and Japan, were less likely than participants from individualistic cultures like in the US to spare children over the elderly. “If you buy a car that was programmed in Germany and drive it to Asia, whose ethics should the car obey?” asks Nallur. The question cannot be answered with more math.
But the more stakeholders involved in the discussion, the better, says Nallur. Physicists are still working to integrate their AI research into the mainstream machine learning community. To understand his role in ethical conversations, Nord says he’s working to partner with social scientists, ethicists, and experts across many disciplines. He wants to have a conversation about what constitutes ethical scientific use for AI algorithms, and what scientists should ask themselves when they use them. “I’m hoping that what I do is productive in a positive way for humanity,” says Nord. As AI applications barrel forward, these physicists are trying to lay the track to a more responsible future.
More Great WIRED Stories
- 📩 Want the latest on tech, science, and more? Sign up for our newsletters!
- “Dr. Phosphine” and the possibility of life on Venus
- Meet this year’s WIRED25: People who are making things better
- How we’ll know the election wasn’t rigged
- Dungeons & Dragons TikTok is Gen Z at its most wholesome
- You have a million tabs open. Here’s how to manage them
- 🏃🏽♀️ Want the best tools to get healthy? Check out our Gear team’s picks for the best fitness trackers, running gear (including shoes and socks), and best headphones
