In 2014, volunteers began hunting for unusual features in particle-collision images; by 2026, over 42,000 had examined nearly 39,000 pictures

Tens of thousands of volunteers with nothing but a laptop and some curiosity spent months combing through snapshots from one of the most expensive machines humanity has ever built, and no physics degree was required to take part. That's the story of a project called Higgs Hunters, launched in November 2014 by physicists working on the ATLAS experiment at the CERN Large Hadron Collider with help from Oxford, NYU, and Birmingham universities. According to a peer-reviewed 2016 study titled “That looks weird' - evaluating citizen scientists' ability to detect unusual features in ATLAS images of LHC collisions," by Alan Barr, Andy Haas, and Christina Kalderon, on arXiv, the effort had, by that point, seen "over 32,000 citizen scientists from 179 countries" spot peculiarities in collision images for the project, logging around 1.2 million points of interest, with numbers growing exponentially as it goes. By the time the project wrapped, it had seen well over 42,000 volunteers from nearly 170 countries spot peculiarities in nearly 39,000 images, per Oxford's own project summary.

The work of spotting the weird stuff

Here's some background. After the discovery of the “God particle,” The Higgs boson, in 2012, physicists wondered what else it could contain. One popular theory was that a Higgs boson could decay into a lighter particle of an unknown kind, which would travel a short distance before decaying itself, producing an 'off-centre vertex' (a point, away from the original collision point, where a particle decays into other particles) in the process. Spotting these by eye turns out to be a task humans have a demonstrable edge at, compared to trying to teach an algorithm to spot all of the odd stuff out.

Which may be why this finding is interesting to readers following debates about AI and human expertise. According to the same arXiv paper, the untrained eyes of these volunteers were "as efficient as the algorithms used by the ATLAS experiment to identify such vertices" in some cases, and beat them in others, for spotting peculiarities in collision data. Crowdsourcing may not replace careful individual work, but it has its own strengths when it comes to finding anomalies. It's also helped spot a few interesting oddities. One image in particular appeared to show a 'muon jet,' a cluster of muons that, if real, would not be predicted by the Standard Model of particle physics. Investigation by the science team found the image showed "punch-through," a known but unusual way a particle can interact with the detector; not a new particle, but still a genuinely rare and interesting detector effect that the volunteers had been right to flag as strange.

More than a science fair

Beyond the purely scientific, the volunteers also seem to have gotten a nice grounding in physics. A companion piece titled “Citizen scientist community engagement with the HiggsHunters project at the Large Hadron Collider,” published in the journal Research for All, also written by Barr, Haas and Kalderon, noted that these volunteers created a technical jargon in the forums for the project, as part of a self-taught process of learning the subject. According to figures published by Oxford, over 80 percent of surveyed volunteers said their knowledge of physics improved during the project's run, and nearly half said they were more likely to pursue further studies in the field. Separately, a project run with the Institute for Research in Schools allowed a group of UK-based teens to contribute findings later reviewed by the CERN ATLAS collaboration.

Why this matters

For a generation raised on solving mysteries and debating the latest headlines, this sort of outreach is old hat. Higgs Hunters was itself one entry in a now well-established tradition of crowdsourced citizen science; the Higgs Hunters project ran on a Zooniverse database, considered the biggest online hub for citizen science. At the same time, it's worth noting that the same questions are being asked about the future of unpaid crowdsourced research.

After all, with every passing year, algorithms and artificial intelligence grow more capable at doing things previously done by humans, but how much does that really matter if it comes at the cost of unpaid and untrained labor? There's no final answer here, but Higgs Hunters is a useful reminder that somewhere between a bored Tuesday night and unpaid curiosity lies the ability for an interested amateur to do something resembling the work of a professional.