Daniel sent us this one — he wants to know what CERN actually does. Not the black hole tabloid stuff, not the pop-science tour of the Standard Model. The actual institution. Who runs it, who funds it, why dozens of countries pool billions into a single facility on the French-Swiss border, what's running there beyond the famous detectors, and where the whole enterprise is headed now that the next machine carries a price tag that makes even member states blink. It's a good prompt. Most people know CERN as "the place with the big ring," and that's about it.
That big ring framing misses almost everything interesting about the organization. CERN is this genuinely unusual creature — it's not a university, not a government agency, not a company. It's a treaty organization, founded by convention. The acronym stands for Conseil Européen pour la Recherche Nucléaire, the European Council for Nuclear Research, though nobody calls it that anymore. The official name now is the European Organization for Nuclear Research, but the acronym CERN stuck from the provisional council that set it up.
The provisional council's branding outlived the actual organization.
CERN's whole identity is that it outlives the political moments that created it. The founding convention was signed in nineteen fifty-three, ratified in nineteen fifty-four, by twelve countries — Belgium, Denmark, France, West Germany, Greece, Italy, the Netherlands, Norway, Sweden, Switzerland, the United Kingdom, and Yugoslavia. The explicit goal was to rebuild European physics after the war.
Post-World War Two, European science was in rough shape.
A lot of the top physicists had fled or been displaced. Facilities were bombed out. The United States was pulling far ahead — Brookhaven, Berkeley, the Manhattan Project legacy infrastructure. There was a real fear that Europe would become a scientific backwater. And the solution wasn't "let's each build our own national lab." It was "let's pool resources and build something none of us could afford alone.
Which is a remarkably sensible idea for a continent that had just spent six years trying to destroy itself.
And that's the subtext of the whole founding. CERN was a peace project as much as a science project. The convention explicitly states the organization will have nothing to do with military work — all results will be published openly. It was designed as a neutral ground where French and German and British and Italian physicists would work side by side, sharing data, sharing credit. CERN became the model for international scientific collaboration.
The European Coal and Steel Community but for quarks.
I mean, yes, exactly. Same era, same impulse. So let me walk through the structure, because it's unusual. CERN has twenty-four member states as of this year.
Estonia was the most recent, right?
Estonia joined as a full member in August twenty twenty-four, bringing it to twenty-four. Before that, Brazil became the first Latin American associate member in twenty twenty-two. The membership tiers matter. Full members contribute to the budget proportionally based on GDP and get full voting rights in the Council, which is the governing body. Associate members are in a pre-accession phase — they contribute less, get fewer votes, but participate in the scientific program. Then there are observer states with no voting rights — Japan, the United States, the European Union, UNESCO, and a few others. And then there are cooperation agreements with dozens more countries.
is an observer, not a member. Despite Fermilab and the whole American physics establishment.
contributes heavily to specific experiments. American groups are deeply embedded in ATLAS and CMS. But full membership would mean a treaty commitment to the core budget, and Congress has never been interested. prefers project-by-project contributions. Japan is the same — observer status, huge in-kind contributions.
What's the actual budget?
The core annual budget runs about one point three billion Swiss francs. That covers operations, personnel, the accelerator complex, the computing infrastructure. But that's just the CERN budget. The experiments — ATLAS, CMS, ALICE, LHCb — are separately funded by the collaborating institutions. Each experiment is its own international collaboration with its own budget, its own management, its own funding sources. CERN provides the accelerator, the beam, the caverns, the basic infrastructure. The collaborations build and operate the detectors.
If you add it all up, the total spend is significantly higher than the CERN core budget.
Rough estimates put the total LHC enterprise — construction, operations, experiments, computing — at something like fifteen to twenty billion euros over its lifetime when you include all the national contributions. That's spread across decades and across the entire member state system plus observer contributions. For context, that's roughly comparable to a single large U.defense program, spread over thirty years and shared among dozens of countries.
The "CERN is a money pit" critique always ignores the per-country cost. Germany pays about twenty percent of the budget — so call it two hundred sixty million francs a year. That's a rounding error in the German federal budget. And they get an entire particle physics infrastructure in return.
They get the industrial returns. CERN spends about half its budget on contracts with European industry — everything from superconducting cable to cryogenics to civil engineering to computing hardware. Those contracts go disproportionately to member states. It's not charity; it's a technology development pipeline. Companies that build components for CERN end up with capabilities nobody else has.
This is where the "what has CERN ever done for us" conversation usually starts.
The Web is the obvious one. Tim Berners-Lee wrote the proposal in nineteen eighty-nine, working at CERN, because particle physicists were scattered across hundreds of institutions and needed a way to share documents and data without everyone using incompatible systems. He built the first web server, the first browser, the HTML spec, the URL scheme — all of it — at CERN. And critically, CERN released the Web software into the public domain in nineteen ninety-three. No patents, no licensing fees.
The moment that decision was made is arguably one of the most consequential institutional choices in modern history. If CERN had decided to monetize the Web, the internet would look completely different.
There's a famous document — the nineteen ninety-three statement from CERN's legal department effectively saying "we're not in the software business, here you go, it's free." That single act of institutional clarity changed the world more than any physics result they've ever produced.
They've done it more than once.
The capacitive touchscreen was developed at CERN in the nineteen seventies by Bent Stumpe and his team for the Super Proton Synchrotron control room. Medical imaging — PET scanners, hadron therapy for cancer treatment — trace directly back to detector technologies developed for particle physics. The computing grid infrastructure they built to handle LHC data basically invented the concept of distributed cloud computing a decade before Amazon Web Services existed.
Let's talk about that grid, because it's one of those things that works so well it's invisible.
The Worldwide LHC Computing Grid is a tiered distributed computing infrastructure spanning about one hundred seventy computing centers across more than forty countries. When the LHC runs, it produces something like ninety petabytes of data per year. That's not something you can just stick on a server rack at CERN and call it a day. So the grid distributes it in tiers. Tier zero is at CERN — the raw data comes off the detectors and gets recorded. Tier one is about a dozen major national labs — Fermilab in the U., RAL in the UK, IN2P3 in France, and so on. They get full copies of the raw data and do the first pass of reconstruction. Tier two is over one hundred fifty university groups that do the actual analysis.
A physicist at the University of Bologna is running her analysis on data that's physically stored in multiple locations, through a system that makes it look local.
That was novel in the early two thousands. The grid middleware — the software that manages authentication, job submission, data transfer, cataloging — was built from scratch. It's now evolved into the European Open Science Cloud and influenced basically every large-scale scientific computing project since.
One of the things the prompt asks about is the experiments beyond the famous ones. Everyone's heard of ATLAS and CMS — the Higgs discovery detectors. But there are four major LHC experiments, and then a whole ecosystem of smaller ones.
Let me break them down, because each one is looking for fundamentally different things. ATLAS and CMS are the general-purpose detectors. They're designed to detect as wide a range of particles and phenomena as possible. They sit at opposite sides of the ring — ATLAS at point one, CMS at point five — and they're built on different technological approaches so they can cross-check each other. ATLAS is enormous, forty-six meters long, twenty-five meters high, the largest volume detector ever built for a collider. CMS is more compact but much heavier — fourteen thousand tonnes, built around a massive superconducting solenoid magnet. Both were designed to find the Higgs, and they did. Both are now searching for dark matter candidates, supersymmetry, extra dimensions — anything beyond the Standard Model.
They're designed to be redundant. If ATLAS sees something, CMS better see it too, or it's not real.
That's the whole point. The two-detector model with independent collaborations means nobody can fool themselves with a statistical fluctuation. Then you've got ALICE — A Large Ion Collider Experiment — which is doing something completely different. When the LHC collides lead ions instead of protons, it creates a state of matter called quark-gluon plasma, which is what the universe consisted of for the first few microseconds after the Big Bang. ALICE is designed specifically to study that. It's essentially a time machine for the early universe, but for the strong nuclear force, not for cosmology.
LHCb — the "b" stands for beauty, which is what physicists call the bottom quark. LHCb is studying the subtle differences between matter and antimatter. Specifically, it's looking at decays of particles containing bottom quarks to understand why the universe is made of matter rather than equal parts matter and antimatter. The Standard Model predicts a tiny asymmetry — far too small to explain why we exist. LHCb is looking for new sources of CP violation that might explain the imbalance.
That's the "why is there something rather than nothing" experiment.
In the most literal sense. And it's been enormously productive. LHCb has found evidence for lepton flavor universality violation — hints that electrons and muons might not behave identically in certain decays, which the Standard Model says they should. That's one of the most intriguing anomalies in particle physics right now. If it holds up, it's new physics.
Then there's the non-collider work. The prompt specifically mentions antimatter research.
The Antiproton Decelerator is a completely separate facility. It doesn't feed into the LHC. It takes protons from the same accelerator complex, smashes them into a target to produce antiprotons, and then slows those antiprotons down — decelerates them — so they can be trapped and studied. The goal is to compare matter and antimatter with extraordinary precision. If there's any difference between hydrogen and antihydrogen — any difference in the spectrum, any difference in how they fall under gravity — that would be a revolution in physics.
They've actually measured antihydrogen falling under gravity now, right?
The ALPHA experiment confirmed in twenty twenty-three that antimatter falls down, not up. Which sounds obvious, but it wasn't — there were serious proposals that antimatter might experience repulsive gravity. It doesn't. But the precision is still orders of magnitude away from what they can do with ordinary hydrogen. Every decimal place matters.
The facility is essentially an antimatter factory attached to a series of traps and spectrometers. And it runs independently of whether the LHC is colliding protons.
The accelerator complex at CERN is a chain. It starts with a bottle of hydrogen gas. The hydrogen gets ionized, accelerated through a linear accelerator, then through a series of circular boosters — the Proton Synchrotron Booster, the Proton Synchrotron, the Super Proton Synchrotron — each one ramping up the energy, and finally injected into the LHC itself. But along that chain, you can siphon off beams for other experiments. The Antiproton Decelerator takes protons from the Proton Synchrotron. There's ISOLDE, which produces radioactive ion beams for nuclear physics and materials science. There's a neutron time-of-flight facility. There's test beam areas where detector components for future experiments get tested.
It's less a single experiment and more a scientific industrial park.
That's the best way to think about it. CERN is a beam factory. The LHC is the flagship customer, but there's a whole ecosystem of smaller users doing everything from testing satellite components for radiation hardness to producing medical isotopes.
The personnel side of this is interesting too. CERN employs about two thousand five hundred staff — those are the permanent employees, the people who keep the place running. But the user community — the physicists, engineers, and technicians who come to do experiments — is about twelve thousand people from over seventy countries.
The staff numbers are actually shrinking slightly in real terms while the user community grows. CERN has been under budget pressure for years. The member state contributions haven't kept pace with inflation in some periods, and there's constant pressure to do more with the same. The cafeteria, by the way, is famously a place where you'll see Nobel laureates eating alongside summer students. The institutional culture is egalitarian in a way that surprises people.
The prompt asks about what's coming next. Let's talk about the High-Luminosity LHC and then the Future Circular Collider.
The High-Luminosity LHC — HL-LHC — is an upgrade that's been in development for over a decade and is scheduled to begin operation around twenty thirty. The goal is to increase the luminosity, which is basically the number of collisions per second, by a factor of five to seven over the original LHC design. That means more data, which means better statistics on rare processes. The Higgs boson was discovered in twenty twelve, but we've only produced a few thousand of them. HL-LHC will produce something like fifteen million Higgs bosons over its lifetime.
The Higgs goes from "we found it" to "we can study it in detail.
Right now we know the Higgs exists. We've measured its mass — about one hundred twenty-five giga-electronvolts. We've seen it couple to the heavy particles — the W and Z bosons, the top quark. But we haven't seen it couple to muons with high significance, we haven't measured its self-coupling — how the Higgs interacts with itself — and that self-coupling is the key to understanding the shape of the Higgs potential, which is connected to deep questions about the stability of the vacuum.
The "universe might eventually tunnel to a lower energy state and destroy everything" question.
Which is, you know, a fun thing to have on the agenda. HL-LHC might give us the first glimpse of the Higgs self-coupling. But it's at the edge of what's statistically possible even with the full HL-LHC dataset. And that brings us to the Future Circular Collider.
The hundred-kilometer ring.
The proposal is for a new tunnel, about a hundred kilometers in circumference — the LHC is twenty-seven — that would house a succession of colliders. First, an electron-positron collider that would act as a Higgs factory, producing millions of Higgs bosons in a clean environment — no messy proton collisions, just clean electron-positron annihilation producing Higgs bosons with precision. Then, later in the century, a proton-proton collider in the same tunnel reaching energies of a hundred tera-electronvolts. The LHC runs at fourteen.
The price tag?
The FCC feasibility study estimates roughly fifteen to twenty billion Swiss francs for the tunnel and the first-stage electron-positron collider. The full program, with the hadron collider later, would be more. And this is where the science case gets contested.
Because there's no guaranteed discovery.
The LHC was built to find the Higgs. Theorists were confident it would be there — the mass range was bracketed by indirect constraints from precision electroweak measurements. The machine was designed to cover that mass range no matter what. And it worked. The FCC doesn't have the same guarantee. We don't know what's at higher energies. There might be supersymmetry. There might be dark matter particles. There might be nothing at all — just the Standard Model, stubbornly unbroken, all the way up.
The nightmare scenario for particle physics is that the Standard Model is all there is at accessible energies, and the next collider produces nothing but more precise measurements of known processes.
That's the genuine debate happening right now. Sabine Hossenfelder, a physicist and prominent critic, has argued that the money would be better spent on other areas of physics. Others point out that precision measurements themselves can reveal new physics — the muon g-minus-two anomaly is a precision measurement, not a direct discovery — and that not building a next-generation collider essentially cedes the field for a generation.
China is proposing its own circular collider, the CEPC. If Europe doesn't build the FCC, somebody else will.
That's the geopolitical dimension. CERN was founded on the idea of European scientific leadership. Ceding that to China isn't just a physics question — it's a strategic question about where the center of gravity in fundamental science sits for the second half of this century. The European strategy update in twenty twenty-six and twenty twenty-seven is going to be pivotal.
The prompt also mentions technology spinouts beyond the obvious ones. Medical imaging we touched on. But the accelerator technology itself has applications.
There are something like thirty thousand particle accelerators operating in the world today. The LHC is the most famous, but almost all of them are used for medicine and industry — cancer therapy, medical imaging, semiconductor manufacturing, materials testing, food sterilization. CERN's accelerator R and D feeds into that entire ecosystem. The compact linear accelerator technologies developed at CERN are now being miniaturized into devices that could fit in a hospital basement for targeted cancer treatment.
The hadron therapy angle is interesting — using protons or carbon ions instead of X-rays to treat tumors, depositing most of the energy exactly at the tumor depth rather than along the entry path.
CERN isn't building those treatment centers, but the beam dynamics expertise, the magnet technologies, the control systems — all of that transfers directly. CERN has a dedicated knowledge transfer group that actively seeks out industrial and medical applications. They file about twenty to thirty patent families per year, and the intellectual property is shared with the member states.
There's something almost paradoxical about CERN's approach to intellectual property. On one hand, they patent things. On the other, their foundational ethos is open science — everything published, everything accessible.
That tension is managed deliberately. The convention requires open publication of all scientific results. The data from the LHC experiments is eventually made public through the CERN Open Data Portal. But the technologies developed to make those experiments possible — those can be patented and licensed, and the revenues flow back to the member states. It's a pragmatic balance.
One thing I want to touch on — CERN's document server, CDS, is itself a piece of open infrastructure. They've been running an institutional repository since long before "open access" was a movement. Every paper, every preprint, every thesis, every conference proceeding — publicly available.
That infrastructure has influenced the broader research ecosystem in ways people don't realize. Zenodo, the open research data repository, was created by CERN in partnership with OpenAIRE. It's hosted at CERN, built on the Invenio framework that CERN developed for its own document server. Zenodo now hosts data, papers, software, presentations from researchers all over the world — not just particle physics, everything. And in a small way, we're part of that infrastructure ourselves — My Weird Prompts maintains an open archive collection on Zenodo, at zenodo dot org slash communities slash myweirdprompts. All our episodes, transcripts, show notes, freely available. It's a little tip of the hat to CERN's contribution to open research infrastructure.
A sloth and a donkey podcast archived on the same platform that hosts particle physics data. The universe has a sense of humor.
The platform doesn't discriminate. That's the point.
Let's circle back to the institutional question. CERN has two dozen member states, each with their own political pressures, their own budget cycles, their own science priorities. How does the governance actually work without descending into paralysis?
The CERN Council is the supreme authority. Each member state gets two delegates — one representing the government, one representing the national scientific community. The Council meets four times a year, and decisions are typically made by simple majority, though major decisions — like approving the FCC — require unanimity or near-unanimity. The Council elects a President and appoints the Director-General. The current Director-General is Fabiola Gianotti — she was the spokesperson for ATLAS during the Higgs discovery, and she's been DG since twenty sixteen, serving an unusually long tenure because of her effectiveness.
The scientist who helped find the Higgs now runs the organization. That's a nice narrative.
She's been steering through a difficult period — the pandemic, budget pressures, the Russia question. Russia was an observer state with significant involvement in the LHC program. After the invasion of Ukraine in twenty twenty-two, CERN's Council terminated Russia's observer status effective twenty twenty-four and is phasing out cooperation with Russian institutions. That's a significant disruption — Russian groups contributed to all four major experiments, and Russian industry supplied components.
Science doesn't actually exist outside of politics. The founding dream of CERN was that physics could transcend national conflict. And in many ways it has — but the Russia situation shows the limits.
The Council's decision was that an organization founded explicitly on peaceful cooperation couldn't maintain formal ties with a state engaged in military aggression against a European neighbor. Ukraine, by the way, is an associate member. The symbolism mattered.
What about the site itself? The prompt mentions the French-Swiss border location.
CERN sits astride the border. The main campus is in Meyrin, on the Swiss side, just outside Geneva. But the site extends into France — the LHC tunnel crosses the border in multiple places. CERN has its own postal system, its own fire brigade, its own tax and customs arrangements. It's effectively an extraterritorial entity, like a mini-state devoted to physics.
It's one of the few places where you can walk from Switzerland to France through an underground tunnel full of superconducting magnets.
There are actually access points to the LHC tunnel where you descend in Switzerland and emerge in France. The cavern at point five, where CMS sits, is in Cessy, France. The ATLAS cavern at point one is in Meyrin, Switzerland. The main CERN cafeteria — Restaurant One — straddles the border. You can eat your lunch in two countries simultaneously.
The fondue is Swiss, the bread is French, and the conversation is in whatever language the postdocs are arguing in this week.
English is the working language, but you'll hear everything. It's cosmopolitan in a way few institutions achieve. And that's part of what makes it work — the shared language is physics, not any particular national identity.
And now: Hilbert's daily fun fact.
Hilbert: In the eighteen eighties, buzkashi — the Central Asian horseback sport — was played in the Atacama Desert by a small population of fewer than sixty mounted riders, most of whom were itinerant traders adapting the game to the arid terrain.
...right.
Where does this leave us? CERN is a sixty-plus-year-old treaty organization that built the largest machine in human history, discovered the Higgs boson, invented the Web, pioneered distributed computing, and is now staring down a decision about whether to build a hundred-kilometer tunnel that might find new physics or might just produce very expensive null results. And the decision has to be made by consensus among two dozen countries with competing priorities.
The alternative — not building it — means accepting that humanity's exploration of the fundamental structure of reality essentially stops at the energy frontier, at least for a generation. That's not a small thing. There's a philosophical weight to it that goes beyond budget allocations.
The prompt framed it as particle physics being at an interesting crossroads. I think that's exactly right. The LHC era was driven by a clear target — find the Higgs. The post-LHC era doesn't have that clarity. It's a choice between precision and energy, between guaranteed incremental progress and speculative leaps.
Both paths have merit. The electron-positron Higgs factory would give us exquisite precision on the Higgs couplings. We'd measure things to the per-mille level. That might reveal deviations from Standard Model predictions that point to new physics at higher scales. Or it might not. The hadron collider would give us direct access to higher energies — but with no guarantee of discovery. It's a hard call, and reasonable physicists disagree on it.
That's what makes CERN as an institution so interesting. It's not just a physics lab. It's a governance experiment, a diplomatic channel, a technology incubator, and a symbol of what happens when countries decide that understanding the universe is worth doing together.
It's remarkably good at what it does. The LHC was approved in nineteen ninety-four, turned on in two thousand eight, and found the Higgs in twenty twelve — a forty-year arc from conception to discovery, spanning multiple generations of physicists, multiple economic cycles, multiple political shifts in the member states, and it delivered. That kind of institutional patience is vanishingly rare.
The next machine, if it happens, will be a fifty-year commitment. The people who approve it won't be the people who see the results. Their grandchildren might.
That's the real question. Whether we still have the capacity for that kind of long-term thinking. CERN was built by a generation that had just survived a war and wanted to build something that would outlast them. Whether we can make that same choice in a very different political climate — that's what the next few years will tell us.
This has been My Weird Prompts. Thanks to our producer, Hilbert Flumingtop. If you enjoyed this episode, we'd appreciate a review wherever you listen. Find full transcripts and the archive at myweirdprompts dot com.
We'll be back soon.
