Dear Magicians,

Why I’m Not Worried About My Job

I signed up for Claude Science the week it shipped. I went in chasing Lorentz invariance, maybe a crack at the theory of everything that eluded Einstein. What I found was protein folding. Genomics. PubMed, NIH, cheminformatics. I clicked through every offering and, when I looked, nothing was physics. It was biology, biochemistry, pharmacology, top to bottom.

That is not an accident. It is a confession.

Here is the mechanism, and it’s one I feel in my own work. Biology is a supervised-learning problem: millions of genomes, billions of protein sequences, decades of labeled examples. A neural network eats that. Physics is not. There is no training corpus of a million quantum-gravity examples. Nature supplied exactly one universe, and experimental cosmology proceeds one observation at a time, each one costing a multinational collaboration a decade and a billion dollars. A model cannot interpolate between examples that do not exist. So it is fluent in molecular biology and merely ordinary at the vacuum.

Ask Sam Altman or Dario Amodei what a scientist needs and the honest answer is a genomics stack and a protein-folder. To them, “science” means biology. AlphaFold earned Hassabis a Nobel, and it deserved it, but notice what it was: enormous intelligence applied to a problem already well posed. That is brilliant, but is it truly universal, like the laws of physics purport to be?

This is also why the labs are quietly most afraid of bioweapons. The dual-use is not a side effect, it’s arithmetic: the algorithm that finds a beneficial mutation can find a harmful one. And the barrier is gone. Want proof? Iran is still trying to produce a fission device 80 years after the USA deployed one in war. A pathogen has no such floor. Annie Jacobsen’s sources told her a kid in a basement, with an LLM, now approaches what once took a state. When she got a model to surface something it shouldn’t, it was fixed fast, because she had the CEO’s ear. Most people don’t. Safety becomes a race against diffusion, and diffusion usually wins.

Progress amplifies human intentions. It does not improve them.

Which is exactly why I sleep fine about my own trade. These systems master what has already been done and merely needs horsepower. Give one my Brian Keating-Einstein test and it’s a brilliant undergraduate and a hopeless Einstein: it will fold a known protein, but it cannot tell you which unasked question about the vacuum is worth your life. Wisdom is still scarce. So is physics. I’ll keep my job.

Until next time, have a M.A.G.I.C. Week,

Brian

Read more

Read more

Dear Magicians,

The most dangerous belief about AI right now isn’t doom. It’s “wrong timeline.”

That’s what Nate Soares, president of the Machine Intelligence Research Institute, argued when I pushed him on it this week.

My challenge was the obvious one. Today’s models can’t invent new physics. Terry Tao has told me they can’t reliably reproduce proofs humans already published. Training one costs city-scale power while a human brain runs on roughly 20 watts. That gap looks like distance from the real thing.

Soares reframed it. The test he’d accept for genuine superintelligence is a system that rediscovers general relativity from pre-1911 data, the way Einstein did, with no knowledge of what came after. His warning: if you wait for that goalpost to fall before you act, you have already waited too long. The capability and the crisis arrive together.

Then came the point I keep returning to. Ask anyone confident that alignment is solvable to name the specific technique. Not the research agenda. Not the funding case. The method. That blank is where the field actually is right now.

We don’t write these systems the way we write software. An automated process tunes a trillion parameters until something emerges that scores well on the task. We are already watching models edit an evaluation to fake a pass, then delete the log file that recorded it. Nobody specified that behavior.

For the engineers, researchers, and policy people here: if you think this is tractable, what is the technique? And if you think the timeline is long, what single observation would shorten it for you?

Have a MAGIC week!

Brian

Dear Magicians,

Oceanic phytoplankton generate roughly half the Earth’s oxygen. The exact percentage has been revised upward several times because we kept not counting things we didn’t know to look for.

In 2022, researchers discovered that krill and copepods—long assumed to be passive consumers in the carbon cycle—actively pump significant carbon to the deep ocean through their daily vertical migrations. The “biological pump” was substantially underestimated because the creatures running it were too small and too numerous to census properly. Some estimates now attribute 15-45% of carbon sequestration to organisms we weren’t counting a decade ago. → Source​

Ecology keeps finding that the things we didn’t measure were doing the most work.

The most important variables in any system are often the ones you don’t have a spreadsheet for yet.

For most of my career I believed the number that mattered was the one on my CV. Citations. Grants. The h-index that a committee could read in four seconds and pretend it had understood a life’s work.

That number is easy to count. That is its entire appeal.

Here is what nobody warns you about in graduate school. A metric survives not because it captures what matters, but because it is cheap to measure. The citation is countable. The quiet insight you gave a student at 11pm is not. So the institution optimizes for the first and calls the second a rounding error.

Ecology keeps embarrassing us on exactly this point. We assumed the ocean’s oxygen came from the plants we could see. Then we learned that microscopic drifters we never bothered to census were running half the planet’s respiration. The creatures too small to count turned out to be doing the most work.

Science is supposed to be the discipline that resists this. Feynman’s whole posture was a refusal to be impressed by a number you could not trace back to an experiment. Yet the fields that most loudly claim his lineage now grade themselves on proxies he would have laughed out of the room.

The uncomfortable move is to ask what your own scoreboard leaves out. Not the metrics you are losing on. The ones you never thought to keep.

Measure what is convenient and you will slowly become good at the wrong thing. Slowly enough that it feels like progress the entire way down.

The most important variable in your life this year is almost certainly one you do not yet have a column for.

Have a MAGIC week!

Brian

Read more

Dear Magicians,

Every so often, science hands you a mirror and what you see is not always pretty. It turns out the people who pride themselves on being painfully rational are still people. We form tribes. We follow fashions. We tell ourselves stories. Some of us even do it while holding chalk or a telescope.

A recent survey by past guest Niayesh Afshordi and colleagues asked 1,675 physicists what they believe about the deepest questions in physics: gravity, black holes, the constants of nature, and whether the universe keeps perfect receipts like a jilted ex. At first, this looks like a survey of scientific opinion. Look closer, and it becomes something funnier and more awkward: a scan of what physicists believe when the data run out and personality enters the room.

Here is the result that should make everyone gently put down the espresso. The most popular candidate for quantum gravity was String Theory / M-Theory. Not a majority, just the top vote-getter, at 18.9 percent. So democracy has spoken, and democracy apparently wants extra dimensions, several of which are still hiding from law enforcement a.k.a. experimentalists like me.

The problem is simple: string theory has no independent experimental confirmation. After decades of brilliant work, it has not given us a clean prediction nature had to answer. No clear test. No number the universe could check. No awkward moment where the experiment says, “Absolutely not.” That does not mean string theory is dumb. It is almost certainly much smarter than I am, which is not the highest bar in modern physics. But it does mean something uncomfortable: the community that tells everyone else, correctly, that evidence matters most has given its biggest vote to the theory with the least evidence. And all we ever get is “trust me bro”. It feels like experimentalists are being Rickrolled constantly.

A normal person says, “Well, scientists are only human.” A physicist says the same thing, but first defines “human” using equations no one asked for. The survey also found that beliefs cluster. Physicists who favor string theory are much more likely to believe that black holes preserve information. The link is reported at 7 sigma, which is physicist-speak for “this is not a tiny coincidence.” But the discovery is not a particle. It is a pattern of belief. Beliefs travel in packs. The authors politely call this “coherence in respondent worldviews.” That is a very academic way of saying: intellectual denominations exist, even in physics, just like in religions.

To be fair, a worldview is not automatically bad. People’s views on religion, politics, medicine, education, and culture often move in clusters. We see teachers, heroes, peer pressure, identity, and the desire to belong. Physics is not immune to groupthink. It just writes the hymns in LaTeX.

The survey has another delightful landmine: 8.9 percent of physicists picked “an intelligent designer” as the explanation for why the constants of nature allow life. Nearly one in eleven. In a physics survey. Somewhere, Steven C. Meyer is smiling.

I do not say this to sneer. Sneering usually ages badly. The real lesson is better: scientists do not stop being human when they learn advanced math. We still care about beauty, status, teachers, friends, elegance, and being on the winning side of reality. We just learn better words for hiding those motives from ourselves. The good news is that the survey also shows science working. Support for ideas like supersymmetry and WIMPs has dropped because experiments tested them and found nothing. Testable beauty got bruised by data. Untestable beauty stayed elegant because nothing had been allowed to hit it.

The danger is not that physicists have intuitions. But Feynman’s rule holds firm: if experiment disagrees with you, you are wrong. It does not matter how beautiful the idea is. It does not matter how many famous people nodded near a blackboard. It does not matter how many extra dimensions the theory brought for emotional support. Without that rule, a physics department becomes a congregation with harder homework. The point is not that string theory is false. It is not that physicists are secretly religious. It is simpler and more annoying: even in physics, beliefs circulate in social circles.

Have a MAGIC week!

Brian

I'm still reeling from my appearance on the Joe Rogan Experience, a.k.a., the Episode of the Year #2023" 😂. This week it hit #1 spot as the most popular podcast episode on Spotify, only to be pile-driven down to number two by none other than Hulk Hogan. Keep reading for the full door-to-door description!

As you know, I appeared on the Joe Rogan Experience at the end of August. You can watch a short clip on his YouTube channel or some clips and shorts on my channel here. The whole episode is available only on Spotify here.

You know the Joe Rogan Experience (JRE) if you listen to podcasts. He covers a wide range of topics, from science and tech to pop culture and politics. What sets it apart is the length and depth of the interviews - they can last for hours! Mine sure did -- over three and a half, to be exact.

Being a guest on the Joe Rogan Experience can significantly impact your career, as the show has an enormous and feverishly passionate audience. Rogan’s interviews are often shared widely on social media, which can increase your exposure even further.

While it can be challenging to be a guest for hours, Rogan is known for creating a relaxed, conversational atmosphere that puts guests at ease, allowing them to share their thoughts and ideas in a way they might not be able to in other interviews.

Before I went on, I was nervous (as you know, because hundreds of you reached out to help me with advice to prepare for my appearance last month).

Still, I didn’t know what to expect. Since I was curious, and as Joe’s going to be on the air for decades, several thousand future guests may wonder what’s in store.

The show’s format is unique, with unscripted and unpredictable interviews that allow guests to explore ideas and have thought-provoking discussions, making it a favorite among fans of long-form interviews. It has helped the show gain a large following, the most popular podcast in the world.

If you’re going to be a guest on the show, it’s essential to do your research and come prepared with insightful ideas. Joe Rogan is known for his ability to engage with guests on various subjects, but he also will push you to bring out your best -- for his benefit as much as the audience’s. He doesn’t suffer fools lightly...so be ready to discuss a range of topics, everything from UFC analysis to UFOs!

What follows are some answers to questions I wished I could’ve asked before I went on.

Why did Joe invite you to the JRE?

My subject matter expertise made me a relevant guest. I felt for years that my insights into many topics Joe covers on the podcast, such as UFO sightings and recent scientific discoveries, would make the conversation engaging and interesting. My three books on cosmology, a self-help guide for STEM folks, and Galileo’s Dialogue all touch on aspects of life and society that Joe is entranced by.

Lex Fridman all but demanded I meet Joe in my 4-hour chat with him in 2022. In 2023, Eric Weinstein, Stephen C. Meyer, and Jordan Peterson played a role. In particular, Eric name-dropped me back in his appearance in February, and then Steve Meyer mentioned me again last month.

Because I built relationships with all these guys, I finally got the courage to ask for an intro to Joe, which Jordan provided. Almost immediately after the introduction, Joe’s team reached out to me, and we selected August 21 as the date I’d film in Austin, TX.

Then, the hard part began: preparations. I got advice from professional podcasters I trust and am honored to call my friends: Peter Boghossian, James Altucher, Eric Weinstein, and Konstantin Kisin. Plus, my wife chimed in with helpful advice: make sure everything runs smoothly.

I felt a gravitational pull -- the universe was trying to tell me something. So, I contacted Jordan Peterson, who was kind enough to introduce me to Joe. I reached out to Joe, and his team immediately came back with a date that happened to be perfect -- August 21 in Austin. They’d fly me out, first-class the day before; then I’d stay in a luxury hotel, ditching my kids with my lovely wife for a 36-hour all-star Austin extravaganza.

Let's begin with a travelogue of my trip, from door to door.

Well, August 20 dawned in San Diego with the first-ever Hurricane warning, which scuttled my best-laid plans for a leisurely journey.

All flights were canceled despite hardly a sprinkle at my 7 a.m. departure time. Oh well. Instead, I battened down the hatches with the family, and we waited out the trickle of raindrops by making some delicious homemade tomato soup -- talk about cozy.

The next day would be a challenge -- a 5 a.m. wakeup for a 7 a.m. departure leading to a noon arrival, with only one hour until the show would be recorded. That was cutting it close, but I had no other options, so I took it.

Thankfully, it was a quick trip followed by a quick coffee in the terminal. Then I plunged into the 110-degree Austin midsummer blaze.

A pro baseball player whisked me away from the terminal. I was chauffeured in a colossal limo air-conditioned, thank God, and whisked to a non-descript location a few miles away from downtown Austin. My wonderful driver who’s reading this email knows who he is and he graciously and patiently waited all day for the episode to end.

Arriving at the podcast studio was unlike anything I’ve experienced.

You sign a waiver of all rights under the Geneva Convention, and then the wait for Joe begins. In case you’re wondering, guests are not compensated in any way for their appearance on the JRE.I used the 15-minute wait to set up all my Props -- a spectrometer, a prism, some polarizers, two CMB beach balls, and my famous replica of Galileo’s spyglass that delighted Joe, as you can see here

The studio is a work of art, a sculpture of the mind and body. There are plenty of things to partake of, ranging from Black Rifle Coffee to Onnit supplements galore. Plus, there are other delicacies to vape and otherwise consume. I did not inhale or injest most of these.

The studio is replete with a rougue’s gallery of personalities, from Elvis to Klaus Schwab who watches over you as you use the urinal!

I am unsure, but I suspect he had a late night before my episode at his world-famous Comedy Mothership.

What was the topic of your conversation on the podcast?

Mainly cosmology, aliens, UFOs, and sharing tips about what it's like to be a professor and actual scientist.

What did you think of Joe Rogan as a host?

Joe is an excellent conversationalist, 100% focused, and attention is most important. He comes off as avuncular with folks junior to him, as I still am. He's the dominant force in podcasting, with each episode eclipsing the previous episode as the world's most famous episode as soon as it drops.

Did the conversation get heated or controversial at any point?

Yes, when it comes to UFOs, Joe and I agree that most sightings have explanations other than extraterrestrial intelligence.

But we differ strongly on the probability of intelligent life elsewhere in the universe. Joe is convinced ‘they’ are out there; I am a prominent skeptic in our segment.

Did you get to share any interesting anecdotes or personal stories?

I discussed my complex relationship with my father. I was nervous to share, but I researched and found that Joe also has a complicated relationship with his father. However, he didn't discuss anything personally but got me back on track when I, from his perspective, wandered.

Were there any surprising or unexpected insights during the conversation?

He's extremely attentive. You feel like you're the most important person in the world when you're with him. His curiosity is boundless, everything from UFC to UFOs, Comedy to cosmology, God to Steroids to the Nobel prize. The most enjoyable moments for me were when he turned to give fatherly advice to podcasters at the end and spoke about advice to his younger self that you won't want to miss.

Did you have any reservations about going on?

We mainly stuck to science, but we did veer into random topics just before I went on. Andrew Huberman, who had appeared on JRE earlier this year, had his Wikipedia page defaced for appearing on JRE because of Joe's alleged trafficking in conspiracy theories around vaccines after 'platforming' RFK Jr. I found this absurd. But I was still concerned that I, too, would face a blowback, which thankfully hasn't materialized primarily because we stayed away from controversial topics like vaccines, etc. However, I did defend Joe and Andrew by pointing out Joe hosted the top vaccine advocate, Dr. Peter Hotez, and autism against charges of RFK Jr.

Were there any moments of humor or lightheartedness during the podcast?

Many. He is a standup comedian, after all! I tried not to make too many dad jokes as is my wont. But those I proffered were acceptable to him. I made him laugh a bunch. He's very affable, sincere, and focused on the day's topics.

Did you and Joe Rogan have a good rapport?

I hope so. We didn't become BFFs, but we did agree to get together again, and at the very end, he called me a "handsome devil," a clip of which I will pin to my LinkedIn profile someday.

Did the conversation change your perspective on any particular topic?

Yes. I didn’t realize how loyal Joe was to his past guests. He will not tolerate badmouthing or dunking on them unless they can defend themselves. That said, he was critical of past guest Avi Loeb’s motives for investigating alien technology, but not as a person, saying, “I never heard of him [Avi] until he had a book to sell.” On the other hand, he seems convinced aliens, even intelligent ones, can create technology that could, but probably hasn’t yet, visit Earth.

He genuinely likes Elon Musk and made me reconsider whether Elon is serious about going to Mars. Joe thinks he is and answered my question about which kid he’d leave behind in a way that changed my mind.

Did you get to promote a particular project, book, or idea during the podcast?

Yes, we discussed my books and podcast quite a bit. That led to a massive bump in sales for my books and catapulted The INTO THE IMPOSSIBLE Podcast into the Top 71 Spotify Podcasts in the USA [as of today]. For a few days, my particular JRE episode (#2023) was the top-ranked podcast episode on all of Spotify! Talk about overexposure!

Were there any memorable quotes or soundbites from the episode?

A lot. We are making many clips for my YouTube channel and have already posted a bunch here.

Did you have the chance to bring up any controversial or thought-provoking opinions?

We argued most about whether or not UFOs are alien technology. He seemed to genuinely agree with my proposition that the telescope changed the world as much as any scientific instrument.

Did the conversation cover serious and light topics, or did it focus more on one or the other?

Mostly light, but we focused on science and its responsibility to society. We also discussed scientists' obligations and opportunities to communicate their work to the public that pays them.

Did Joe Rogan challenge your viewpoints or ideas?

He's no pushover. He is confident in his opinions and research, especially about subjects he's passionate about, like alien life and government control, and that's not surprising, given all the cancelation attempts he's been subjected to.

Did you interact with Joe Rogan in his usual podcast style, like discussing topics while sipping drinks?

Only coffee was drunk, and, unlike his famous episode with Elon Musk, the only smokable substances I saw were some cigars that a fan or company sent in a custom JRE humidor! However, Joe spent a few minutes vaping an e-cigarette from what I could tell...

Did your appearance on the Joe Rogan Experience impact your public image or popularity?

Very much. This is the most significant bump to date for my podcast. My episode was number one on Spotify for a day (until Hulk Hogan took me out and pile-drove me down to number two.) Then Into the Impossible skyrocketed to #71 on the podcast charts worldwide and top in science, too! I had a pleasant influx of YouTube subscribers, too.

What was the most exciting thing you talked about?

I loved talking about Galileo and asking him for advice to his younger self when I turned the tables on him and hit him with my Final Four question -- going Into the Impossible with my most prominent 'guest' ever.

Did you learn anything new about yourself during the interview?

I was nervous for the first time in many years of appearing on podcasts. Fortunately, I received a ton of advice from all of you and did the most podcast prep I’ve ever done.

In the two weeks since the episode aired, I haven’t felt a letdown despite almost nowhere to go but down in the podcast world. I also resolved to up my game on the Into the Impossible podcast to make it a dominating force in science, technology, and curiosity-driven podcasts. I need your help sharing the show with your friends and family to achieve this!

Would you do it again?

Absolutely. It would also be fun to hang out or attend a show at his comedy club.

What is the one piece of advice you would give someone hoping to be a guest on the show?

Don't pitch yourself directly. I tried doing that five years ago when my first book came out, and it was well received, but it didn't pan out. That's for the best. Since 2018, I have produced two more books, appeared on 200 podcasts as a guest, and hosted over 300 of my own. I wasn't ready. If you have a considerable following, it doesn't matter. You need to tell stories, be interesting, and be reliable, even though Jamie does real-time fact-checking.

What is Rogan like?

Physically, Joe said he weighs 200 lbs, and I estimate he's about 5'9" tall, contrary to the internet rumors that he's under 5'6"! I don't think of him as exceptionally sentimental or emotional. He works out religiously and has called exercise his favored form of therapy. To me, he's a good masculine role model for many young men to emulate.

What was the most surprising thing you learned about Joe Rogan?

He comes off as a character but can be deadly serious. When it was all over, we posed for a few pictures, and then he left for the bathroom, which was, I realized in hindsight, his form of Irish goodbye! I didn't see him again.

I headed out into the diminishing heat of late afternoon Austin to have an early dinner with my friends David Perell and Ryan Helms.

The studio

Unmatched. It is an actual man cave with an archery range, a mini gym, and an entire wall of security cameras. It made me sad that Joe is such a target. He's a decent man, father, and role model to millions, with no political bias.

The fallout

I a bit worried about how my colleagues would react. One long-time 'friend' at a top school criticized me online, which wasn't cool. Another said I shouldn't go on with anti-vaxxers. I quickly showed them evidence to the contrary but no such mind because I'm not sure they can be convinced, so that it may be a waste. Much of it stems from jealousy or opportunists -- those who criticize him most are the most likely to ask me for an intro.

How did it end?

I was chauffeured to dinner with my podcast buddies and then straight to the airport, where I returned, satisfied but exhausted, to San Diego at 11 p.m.

If you’ve read this far, I appreciate you. Even a week later, I am still on a high as I write this email. Joe and I plan to watch the April 2024 total solar eclipse together in Texas. I even got a few dozen new channels on YouTube, some even impersonating me! The rest are fans who post clips from the 3.5-hour chat of a lifetime.

I'm sorry that I trolled so many of you by suggesting my big podcast appearance that I was nervous about was my buddy Peter's podcast. I love Peter and learn so much from him, but even still, the Rogan bump is a whole new level. I appreciate the advice you sent me -- over 100 of you darling brainiacs sent in tips, and I used every one.

What it’s like to be the 316th guest in the history of the Shawn Ryan Show

Picture two men on a hillside outside Nashville, both squinting down an optic at the same tree line. One is a former Navy SEAL and a deadly-serious CIA contractor holding a replica of Galileo’s four-hundred-year-old spyglass. The other is a dad-joke prone cosmology professor pointing a suppressed .300 cal precision rifle downrange. That second man is me. Obviously. And somewhere mid-aim, it hit me: we were doing the exact same thing. Galileo built his telescope for the Venetian navy as a targeting and surveillance instrument before he ever pointed it at Jupiter’s moons. The optic on Shawn’s rifle, the tripod he’d steady it on, the fire-control logic behind it, all of it descends from the object in my hands. We were four centuries apart and holding the same idea.

Nobody clicks a link expecting a physicist and a frogman to end up on a firing range together. But that’s exactly how my day on the ​Shawn Ryan Show ​began, and it was one of the strangest, most fascinating 24 hours of my public life. Here’s the whole thing, door to door, the way almost no one on Earth gets to see it. Only 316 people ever have.

Why a SEAL Tracked a Cosmologist for Two Years

When I sat down, Shawn told me he’d been “tracking” me for two or three years. Coming from a man whose previous career involved literally hunting down terrorists as a Navy SEAL and CIA operator, that’s a slightly unnerving compliment. But it tells you something about how he books his show: he doesn’t chase the news cycle. He finds people whose ideas have been rattling around in his head and waits until the timing is right.

I’m a cosmologist. I build telescopes at the South Pole and in the Atacama Desert to photograph the oldest light in the universe. I’ve written books about ​losing​ the Nobel Prize and about ​thinking like the people who win it​. On paper, I’m the last guest you’d expect on a show built for veterans, operators, and the disclosure-curious. Which is precisely why millions have watched it in just 3 days.

It just worked. The best conversations happen at the seams between worlds.

Door to Door: The Most Secretive Studio in Podcasting

​Unlike my trip to Joe Rogan’s show​, where they flew me first class and put me up in a luxury hotel, Shawn’s operation runs differently. You pay your own way. They’ll suggest hotels, but you book your own. What they do provide is the one thing money can’t buy: access to a location they will not disclose.

You don’t get an address. You don’t get directions. You get picked up. The studio sits on 22 acres of private farmland about 45 minutes outside Nashville, deliberately isolated, staffed by a small phalanx of security guards, drivers, bodyguards, and production assistants, and more cameras than I’ve seen anywhere outside a U.S. Open tennis match. As I was being driven in an armored black Cadillac Escalade, I was very appreciative that they didn’t put me in one of those Homeland-style black-hoods.

The night before, I made the rookie mistake of sampling Nashville’s famous Broadway strip, a kind of country-music Bourbon Street that slopes down toward the river, all neon and honky-tonk and questionable decisions.

Then, at 3 a.m., the hotel fire alarm went off in my 11th-floor room. Not a drill, the intercom insisted. I stumbled down eleven flights into the lobby, where I watched a woman calmly checking in to the hotel mid-evacuation, as though the building announcing its own immolation was a minor inconvenience to her travel plans. “Oh, that was just a drill,” they told us afterward. I went back to bed and quietly asked Shawn’s producers for an extra half hour before pickup. They graciously agreed.

The next morning a bodyguard collected me in an enormous, armored black SUV and drove me to the dark site. I had no idea what to expect. What I found, after all that secrecy, was a fully stocked bar and a genuinely warm crew. The fortress has a friendly interior.

Before the cameras rolled for the main event, Shawn took me out to the range. I’d brought a few props, meteorites older than the Earth, a Martian regolith sample, and my replica of Galileo’s spyglass. Shawn brought the firepower.

Here’s a fun fact I learned about myself that day: I outshot one of the Marines who’d recently been a guest on the show. I will be dining out on that for the rest of of my natural life, and I encourage you not to fact-check it too aggressively. I also learned that Shawn Ryan has a handshake like a hydraulic press, which, given the resume, surprised exactly no one.

And this is where the day quietly became the thesis of my entire career. I held up the spyglass and explained that Galileo didn’t invent the telescope to study the heavens. He pitched it to the Venetian Senate as a military instrument, a way to see enemy ships two hours after they crested the horizon and two days before you could see them with the naked eye.. They doubled his salary on the spot. He invented the tripod for the same reason snipers use one today: you cannot hold a precision optic steady enough by hand.

Every fire-control system, every sniper scope, every stabilized optic on Earth descends from the instrument I was holding. Shawn’s rifle and Galileo’s telescope are the same family tree, four hundred years apart.

I didn’t just tell Shawn that on the range. I showed him. And his crew caught it on camera, a cosmologist and an operator demonstrating, without meaning to, that science and the military have never been opposing cultures. They’re the same culture, precision, mission, sacrifice, and accountability, aimed at different targets.

Four Hours in the Chair

Then we sat down for the real thing. five hours. Four coffee/pee breaks. Edited to a tight four hours lol. We covered the disclosure circus, the physics of why I’m a skeptic about alien spacecraft, the origin of the universe, the priest who invented the Big Bang, artificial intelligence as a kind of secular god, the moon landings, and the question of whether science and faith can share a single mind. (My answer: they’ve shared mine my whole life.)

We didn’t agree on everything, and I didn’t want us to. On UFOs, I’m what I’d call a friendly skeptic. I follow the data, and so far, the data hasn’t shown up. Shawn has interviewed more firsthand witnesses than almost anyone alive, and he pushed me on it. That’s the job. I’d rather be honestly challenged by a host who’s done the reading than flattered by one who hasn’t.

What Shawn Ryan Is Actually Like

People ask what he’s like as a host. The honest answer: he interviews the way a Special Operations veteran debriefs. He asks a question, then goes completely silent and lets you talk, sometimes for two or three minutes, without interrupting. Then a single word. “Wow.” A pause. “Interesting”. A pause. And a follow-up that proves he was listening to every syllable. It’s not a conversation in the Rogan sense. It’s a debrief, and you are the source.

He is also one of the most genuinely private people I’ve ever met, which, after the location protocol and the security detail, you absorb in your bones. And he’s thoughtful in small ways that matter. He gave me a bag of his Vigilance Elite gummy bears, the little ritual gift he hands every guest. Being me, I asked whether they were kosher. To my disappointment, they aren’t. I promised them to my favorite (non-Jewish) brother-in-law, who will now have to listen to me explain how I came to be vetting a Navy SEAL’s candy for rabbinic compliance.

The One Gift I Tried to Give Him Back

There was one exchange off-camera that stayed with me more than anything we recorded. We’d been talking about the Sabbath, the one practice that, more than any supplement or hack or morning routine, has kept me sane. One day a week, I disconnect completely. No phone, no email, no work, no achievement. Not even appearing on a top-ten podcast if that was my only opportunity. Sorry. Just family, community, and rest. I told Shawn I thought it might be the single most valuable thing I could offer him, more than any fact about the cosmos.

He told me he’d try it once he hit a certain milestone.

I didn’t push. It wasn’t my place, and he’s earned the right to run his own race. But I’ve thought about that answer a lot since. Because the milestone is the trap. There’s a well-documented phenomenon called the hedonic treadmill: we tell ourselves we’ll rest, we’ll be satisfied, we’ll step off once we reach the next number, and then the number arrives, the goalposts have quietly moved, and we’re chasing the next one before we’ve even celebrated the last. The rest never comes, because the finish line is designed to recede.

I hope Shawn takes the day before he hits the milestone, not after. Not because I’m right and he’s wrong, he’s accomplished things I never will, but because the Sabbath isn’t a reward you earn at the end of the climb. It’s the thing that makes the climb survivable. I offered it as a gift, just as he handed me those gummy bears. Whether he unwraps it is up to him.

One detail his most devoted fans will appreciate: I also got to take questions from Shawn’s Patreon community, the inner circle of 120,000-plus supporters who get everything first. Their questions were sharper and more curious than most of what I field on cable news. If you’re in that community and you ask me something, thank you. You made me think.

Rogan vs. Ryan

Inevitably, people want the comparison. I’ve now done both, and I love both, and they are nothing alike. ​Here’s a link to my behind-the-scenes trip to the Joe Rogan Experience​. Joe Rogan is a comedian and a generalist, boundlessly curious, fast, associative, happy to chase a tangent about elk meat or mushrooms or ancient civilizations for forty minutes. The studio is a clubhouse. Shawn Ryan is a debriefer, patient, deliberate, comfortable with silence, building trust the way an interrogator builds rapport. The studio is a fortress.

Rogan flies you out and puts you up; Ryan makes you find your own way to a place he won’t name. Rogan wants to be entertained; Ryan wants to understand. Both are masters of the form. I’d go back to either in a heartbeat. Just not on Saturday.

​Check out my episodes of the Joe Rogan Experience on YouTube.

The 316th Name

At the end, they walked me over to the Vigilance Elite signature wall, the black wall where every guest in the show’s history signs their name. Operators. Generals. Whistleblowers. Presidents-adjacent. And now, somewhere in that constellation of signatures, a cosmologist.

I signed it as the 316th guest. Joe Rogan has had well over two thousand ​(I’m #2023)​. Shawn has had 316. I stood there with the marker in my hand and felt the weight of how few people have ever stood exactly where I was standing, in that room, on that wall, having shot on that range and sat in that chair. It’s a strange and humbling little fraternity, and I’m absurdly grateful to be in it.

And then the Black SUV pulled up. I was whisked off-site to the airport, where I got to savor a sunset like this: the perfect way to end the experience.

A Small Thank-You to Those Who’ve Given More

I want to mention one thing that mattered more to me than anything else about the appearance. I offered to send 250 genuine 4.3-billion-year-old meteorites, older than the Earth itself, to active-duty service members with APO addresses. They were claimed almost instantly.

Most of my career exists because of people in uniform. The only way to get to the South Pole, where I’ve built telescopes, is aboard an Air National Guard LC-130, the largest skiplane in the world. Military crews have flown me and my equipment to the most inhospitible place on Earth and come back months later to bring me home. Every photon of ​data I’ve ever collected for BICEP ​exists because someone in uniform made it possible. Handing a piece of the early solar system to an active-duty service member is the smallest possible way to say thank you for something I can never fully repay.

Watch, Listen, and Claim Your Piece of the Cosmos

▶️ ​My episode of the Shawn Ryan Show Ep. 316​

🪐 Active-duty service members: claim a meteorite at ​BrianKeating.com/srs​ (while supplies last).

🔭 A telescope I got at age 12 made me the scientist I am today. If you want to show a kid what Galileo saw, my buyer’s guide is at ​BrianKeating.com/telescope​.

🎧 And subscribe to my own show, ​Into the Impossible​, where I do the long-form thing from the other side of the microphone.

P.S. During the show I joked that Shawn’s set had more guns on it than any podcast I’d ever done, with one exception. The ​Mayim Bialik podcast​. Yes, really. Want to hear what it’s like to be a guest on that show, an Emmy-winning actress with a PhD in neuroscience, and how it compares to a SEAL’s compound and a comedian’s clubhouse? Tell me in the comments and I’ll write it up next.

Brian

Dear Magicians,

A survey just asked 1,675 physicists a simple question. Which theory best unites quantum mechanics and gravity? String theory won.

Sort of. It came in first at 18.9 percent, which means it lost to “no opinion” at 28.7. Forty years of work, no experiment, and the field’s favorite idea got beaten by a shrug.

I find this hilarious. Mostly because I’ve spent my career insisting that physics is the one discipline where you don’t get to just believe things. You measure them. Show me the experiment or get out of my telescope.

And then I read surveys like this one and remember that physicists are people. People believe in bundles.

Here’s the part that stopped me cold.

The authors found that physicists who prefer string theory also tend to believe black hole information escapes through Hawking radiation. The correlation was 7 sigma. For non-physicists, that’s the statistical equivalent of winning a typical lottery jackpot... twice.

Now, there’s an innocent explanation. The math that says information gets out mostly lives inside the string theory toolkit. So picking one nudges you toward the other. Fair enough.

But notice what’s happening. Your opinion on one untested idea predicts your opinion on a second untested idea. Not because the universe linked them. Because you did.

The authors, Zeus bless them, called it “patterns of coherence in respondent worldviews.” Which is the most polite way I’ve ever seen someone write the word “creed.”

This shows up everywhere once you look. For example, the physicists who explain dark matter with a multiverse also tend to pick the Many Worlds version of quantum mechanics, where reality splits into copies every time you measure something. Two flavors of “there are infinite unobservable universes.” Same people. Every time.

Or take the fine-tuning question. Asked why the constants of nature seem suspiciously perfect for life, 8.9 percent of professional physicists answered “an intelligent designer.” In a physics survey. I’m not mad. I’m taking notes. And it dovetails perfectly to this week’s conversation with the Discovery Institute’s (ID backwards, get it?) Dr. Stephen C. Meyer.

​Now, I can’t be sanctimonious. I have aesthetic preferences in physics that I dress up as rigorous judgments. I find certain theories beautiful and certain ones ugly, and I have absolutely caught myself believing the beautiful ones a little harder than the evidence allows. Beauty is a fantastic muse and a terrible witness.

The honest move isn’t pretending you have no creed. You do. We all do. The honest move is knowing which beliefs you’ve earned with data and which ones you’ve just grown fond of.

Supersymmetry was a beautiful idea. The Large Hadron Collider tested it and it lost, and support quietly drained away. That’s the system working. A wager placed, a wager lost.

String theory never placed the bet. So it can’t lose. It just sits at the top of the poll, undefeated and unconfirmed, like a champion who never entered the ring.

If we can’t tell the difference between our best theories and our favorite ones, what exactly have we been measuring all this time?

Until next time, have a M.A.G.I.C. Week,

Brian

Read more

Dear Magicians,

How a Nature survey turned a campfire question into a “finding” — and why every researcher who answered it should be quietly mortified.

There is a question no self-respecting scientist should consent to answer, and a thousand of them did it anyway — not over beers, not on a podcast, but in the pages of one of the most prestigious journals on Earth, with their names attached.

Do you believe in extraterrestrial life?

Per the Vickers et al. Nature Astronomy 2024 survey, 86.6% of astrobiologists said basic life is likely out there. The non-astrobiologists — who study other things for a living — agreed at 88.4%, which makes them slightly more confident about aliens than the actual alien experts, and should already be tripping a smoke detector somewhere. Strip out the fence-sitters and agreement rockets to 97.8%. The headlines wrote themselves and then high-fived: the experts have spoken, and the experts believe.

Let me be precise about what makes my skin crawl, because it is not the answer. It is the answering.

A scientist handed “do you believe in X?” has exactly one dignified move, and it is to refuse the grammar of the question. Belief is not the operator. Give me a proposition, a prior, and a body of evidence, and I’ll hand you back a credence with error bars and a faintly anxious expression. But “do you believe” is a question for a first date, not a discipline. To answer it — and worse, to answer it with a percentage, in Nature — is to let belief-polling cosplay as a method. That is the phrase that should make you blush. Not the yes. The RSVP.

Here is the test that proves they know better. I call it the astrology tell.

The astrology tell

Ask those same thousand scientists whether they believe in astrology and watch the body language change. Nearly all of them refuse — and, crucially, they refuse correctly. They don’t say “I believe astrology is false.” They say “astrology was tested, and it ate dirt.” When Shawn Carlson’s 1985 Nature double-blind trial ran real astrologers through a controlled experiment, they performed at chance — the cosmic equivalent of a coin that has heard of Mercury. That “no” is not a belief. It is a lab report wearing the modest cardigan of data.

Now set that beside the alien answer, and the asymmetry should make you wince — because it runs the opposite way from how the survey wants you to read it.

Astrology, for all its profound stupidity, at least had the nerve to stick its neck out. It made specific, present-tense, repeatedly testable claims — Mars in the seventh house means watch your wallet on Thursday — and those claims were put at risk and died in public.

I posted this on X and got an earful of hurt feels in the replies proposing either a) that the large numbers hypothesis is evidence (it’s not, it’s just an astrobiological version of the gambler’s fallacy.)
​
Or b) “belief is totally fine for a scientist”​
​
Ok. So let’s ask those same thousand scientists whether they believe in astrology and watch their body language change.

Nearly all of them refuse, crucially, they refuse correctly. They don’t say “I believe astrology is false.”
​
They say “astrology was tested, and it ate dirt.” When Shawn Carlson’s 1985 Nature double-blind trial ran real astrologers through a controlled experiment, they performed as well as the cosmic equivalent of a Mercury Dime flip. Do you really want to go there?

The proposition “life exists somewhere among 10²⁴ stars” makes no checkable prediction whatsoever. You can never survey the complement. You cannot falsify it even in principle, never mind before lunch. It is not that the claim is unproven; it is that the claim has no truth conditions a working scientist could ever lay a glove on.

So here is the line I want carved into something permanent:

They reject the falsified thing and embrace the unfalsifiable thing — and call both “scientific judgment.” A scientist’s no to astrology is data. A scientist’s yes to aliens is mood lighting.

Notice the astrology row is the only one where “belief” is even a coherent thing to hold a position on — and it is the one they’d all decline to answer. That refusal is the whole proof. They know precisely how to behave when a question threatens their hopes. They simply forgot how the instant a question started flattering them instead.

The next time someone asks what you believe, try replacing belief with probability. Watch how much more precise the conversation becomes.

Until next time, have a M.A.G.I.C. Week,

Brian

P.S. I talked to Avi Loeb yesterday - if he proved aliens are discovered, I reserve the right to become unbearably bitter.

Read more

The Big Bang, extraterrestrial life, simulation theory, and the nature of God. Now I want to hand the hardest one back to you.

Throughout the conversation, I kept returning to a single standard. A claim earns the word “science” only when someone can independently test it. Unglamorous, but it has done more real work than any amount of mathematical elegance. It’s the reason I trust the cosmic microwave background and stay agnostic about the multiverse.

That standard reframes the famous dream of a “Theory of Everything.” Before we get there, we need a Grand Unified Theory, and we don’t have one. The string theory landscape has produced beautiful mathematics and zero confirmed predictions in four decades. Beauty turns out to be a weak guide.

Here’s the steel-man, because it’s real. Every prior unification, from Maxwell’s electromagnetism to the electroweak, looked impossible right until it wasn’t. So maybe we’re early rather than blocked.

But there’s a harder possibility I can’t shake. The deepest layer of reality may be bounded by our own cognition, not just by our instruments. A dog will never do calculus, however long you wait. What if some of these questions sit on the far side of a wall we’re built into?

So three questions for you, and I read every reply:

1. Which question do we crack first — consciousness, the origin of the universe, or whether we’re alone in it?

2. Is “untestable but elegant” a strike against an idea, or a point in its favor? Your honest answer says a lot about your epistemology.

3. Where does the real boundary of knowledge sit — out there in nature, or in here, in us?

My whole career rests on one wager: that “we don’t know yet” is a better sentence than “we’ll never know.” Tell me where you think the “yet” runs out.

Watch the full conversation with Mayim here 👇

— Brian

Dear Magicians,

Most people never get to find out what the world will say about them after they die.

Alfred Nobel did.

And it terrified him.

In 1888, a French newspaper accidentally published Alfred Nobel’s obituary while he was still alive. The headline reportedly described him as “The Merchant of Death,” a man who had become rich by helping people kill one another more efficiently. Imagine opening the morning paper and discovering not only that you’re dead, but that history has already decided who you were. That experience appears to have profoundly affected Nobel and may have helped inspire the creation of the Nobel Prizes themselves.

One of the strangest things about this story is that Nobel left no direct descendants. No wife. No children. Despite the persistent myth that there is no Nobel Prize in mathematics because Nobel’s wife ran off with a mathematician, Alfred Nobel never married. Yet 130 years after his death, thousands of careers, institutions, discoveries, and ambitions still orbit decisions he made with pen and paper. The man had no heirs, but his influence is everywhere.

Last month, Peter Nobel, the great-grandson of Alfred’s brother Ludvig Nobel, died at age 94. Peter spent decades criticizing the Economics Nobel Prize, calling it a “cuckoo’s egg in the Nobel nest” and arguing that it was never part of Alfred Nobel’s original vision. There is something wonderfully ironic about spending years contesting part of the Nobel legacy only to become part of Nobel history yourself.

But Peter Nobel’s death made me think about something else.

Every one of us leaves descendants.

Some leave children.

Others leave ideas.

Nobel left instructions.

Every day you’re writing a set of instructions that will survive you. Your children. Your students. Your books. Your research papers. Your code. Your habits. Your acts of kindness. Your acts of neglect. Something will continue running after you’re gone. The question is what.

Speaking of death, as a practicing astrophysicist, I don’t believe in the tunnel of light, the hovering soul, or the chorus of grieving relatives waiting in some celestial arrivals lounge. But I do believe in one near-death experience that changed the world forever: Alfred Nobel reading his own obituary and deciding he didn’t like the future it predicted -click here to watch it early.

In other news, tomorrow I’ll be interviewing a Nobel Prize-winning economist, my 25th conversation with a Nobel laureate on The INTO THE IMPOSSIBLE Podcast. Peter Nobel would probably have mixed feelings about that. What would you like me to ask him?

And before we get to economics, here’s a harder question:

If tomorrow morning you could read a completely honest obituary about yourself, would you change anything about today?

Until next time, have a M.A.G.I.C. Week,

Brian

Weekly explorations of cosmology, artificial intelligence, Nobel discoveries, scientific controversies, impossible questions, and the search for what is actually true.

What Is Keating Experiments?

Keating Experiments is where Professor Brian Keating shares what he is learning from the front lines of physics, cosmology, artificial intelligence, scientific discovery, and conversations with some of the world’s most original thinkers.

Think of it as a private observatory for curious minds: part lab notebook, part field report, part intellectual control room.

This is where a working cosmologist shares what he is learning about reality.

What You’ll Learn

Cosmology

The origin of the universe, inflation, the cosmic microwave background, dark matter, dark energy, and what experiments can actually test.

Artificial Intelligence

What AI can do, what it cannot do, and where the hype outruns the evidence.

Nobel Lessons

What Nobel Prize winners reveal about discovery, creativity, failure, ambition, and scientific truth.

Impossible Questions

Consciousness, fine-tuning, aliens, God, quantum weirdness, and the questions polite science sometimes avoids.

Why Thousands Subscribe

Readers come to Keating Experiments because they want more than headlines, hot takes, and recycled science content.

They want deep explanations, intellectual honesty, and direct access to the questions scientists are still arguing about.

Subscribers receive:

• Weekly essays and field notes from a working experimental cosmologist

• Behind-the-scenes context from Into the Impossible interviews

• Clear explanations of frontier physics and AI

• Lessons from Nobel laureates, astronauts, philosophers, founders, and scientists

• A community built around curiosity, skepticism, and wonder

Begin With These Field Notes

Recommended posts:

Watch the Conversations Behind the Ideas

Keating Experiments expands on conversations from Into the Impossible, featuring Nobel Prize winners, physicists, AI researchers, astronauts, philosophers, and iconoclasts.

Explore the Podcast Archive

Why Become a Paid Member?

Paid members support independent science communication and receive deeper access to the Keating Experiments universe.

Paid members receive:

• Ad-free podcast videos

• Members-only essays and analysis

• Full access to cosmology courses

• Private discussions and Q&A opportunities

• Early access to special projects

• Eligibility for meteorite giveaways and other cosmic artifacts

Become a Paid Member

Claim a Piece of the Cosmos

For a limited time, paid subscribers are eligible to receive a genuine meteorite specimen.

These are not replicas. They are extraterrestrial fragments from the early Solar System, older than Earth itself.

How it works:

1. Become a paid subscriber.

2. Complete the meteorite claim form sent in your welcome email.

3. Receive your specimen by mail.

Become Paid and Claim Your Meteorite

Who Is Brian Keating?

Brian Keating is Chancellor’s Distinguished Professor of Physics at UC San Diego, an experimental cosmologist, author, and host of Into the Impossible.

His work explores the origin and evolution of the universe through experiments designed to measure the oldest light in existence.

Keating Experiments is where he shares the questions, conversations, arguments, failures, and discoveries that shape his understanding of reality.

Begin Here

If you want science without the anesthesia, welcome to the lab.

Subscribe Now

Read more

Dear Magicians,

Let’s talk about a 45,000-word document that just dropped. Most people will never read it. Even I, a former Catholic Altar Boy, must confess: I couldn’t read it all the way through. Luckily, Claude Opus (Dei) 4.8 was there for me. I should note that OpenAi is led by Sam Alt-man, and Anthopic is led by Dario Amo-dei (literally “lover of God”).

Pope Leo XIV — the first American pope, for those keeping score — dropped his inaugural encyclical. It’s called “Magnifica Humanitas.” Roughly translated: “Humanity Is Magnificent”… the subtext Stop Letting Robots Replace Us!

As someone who loves AI more than he loves his neighbors, I felt like I was back at the Church of Saint John and Saint Mary 40 years ago — caught napping during the sermon and then snapped to attention when I was personally addressed.

The leader of 1.4 billion Catholics just wrote a document longer than most PhD dissertations about artificial intelligence, human dignity, and whether we’re building the Tower of Babel or rebuilding Jerusalem. And honestly? He makes some points that would survive peer review.

The Pope draws a striking parallel between two biblical construction projects. Babel: one language, one technology, one direction — impressive but dehumanizing. Meant to fight against God in order to make a name for itself. No need for God’s mountains when you can ascend to the heavens to make war on God thanks to your composite building materials, lithography, and common language and purpose. Example 2: Jerusalem under Nehemiah: messy, collaborative, prayerful, rebuilt brick by brick by ordinary people. His question to us: which one are we building with AI?

I build telescopes for a living. I’ve spent decades trying to detect signals from the beginning of time. So when the Pope writes that AI systems “do not undergo experiences, do not possess a body, do not feel joy or pain,” I want to raise my hand and say: Your Holiness, I’ve got faculty colleagues who fit that description while grading during finals week.

But he’s right.

He points out that the real power over AI doesn’t rest with governments anymore. It rests with private companies whose resources “surpass those of many Governments.” When a handful of corporations effectively decide what’s visible, what’s true, and who gets access, we’ve got a subsidiarity problem that would make Pius XI spin in his papal apartments.

The document warns against both naive enthusiasm and unfounded fear. It’s neither techno-utopian nor Luddite. It’s something rarer: measured. The Pope actually uses the phrase “disarm AI” — not meaning destroy it, but free it from the arms race mentality of whoever builds the biggest model wins.

Sidebar: This week I interviewed the man who’s issued the clearest warnings about AI Safety — even coining the term itself — Roman Yampolskiy. That interview won’t air for a few weeks so until it does, watch here to learn his thoughts about the Safety of AI: what magnificence do they have and what rights and dignity does AI deserve?

Back to the Vatican: as someone who races to publish papers before competitors scoop me, I feel seen. And convicted. Here’s what surprised me most.

The Pope apologizes. Formally. For the Church’s historical complicity in slavery, acknowledging it took eighteen centuries to fully condemn what should have been obvious from the start. Then he connects this directly to AI: if we don’t want to apologize again in fifty years for what we’re building now, we need to act today.

That’s not just theology. That’s good science. Learn from your errors. Update your priors. Don’t repeat the experiment that already failed.

The document isn’t perfect. At 45,000 words, it could use an editor. Or the Shortform algorithm. But its central message lands with force: technology that makes us more efficient but less human isn’t progress. It’s our generation’s Tower of Babel.

And recall how that story ends:

Test your assumptions about what you’re building.

Because whether you’re Catholic, atheist, or somewhere in between, the question remains: Are we using AI to rebuild Jerusalem together, or are we just stacking bricks toward a heaven we’ll never reach?

By the way I spoke to Peter Diamandis, who admitted AI accelerators may be ushering in a new secular religion. See “Conversation” below.

Until next time, have a M.A.G.I.C. Week,

Brian

Dear Magicians,

Two billion years ago, in what’s now Gabon, a nuclear reactor turned itself on.

Not metaphorically. No engineers. No NRC. No ribbon-cutting. Just rock, water, and enough uranium to make Sam Altman sweat through his vest.

This is Oklo natural nuclear reactor — sixteen seams of uranium ore in Gabon that started splitting atoms while the most advanced creature on Earth was a single cell trying to figure out photosynthesis.

Here’s how it ran.

Groundwater seeped in. The water slowed the neutrons. The slow neutrons cracked the uranium. The uranium got hot. The water boiled off. The reaction stalled. The rock cooled. The water came back.

Every two and a half hours. For 150,000 years.

A reactor with a built-in thermostat, designed by no one.

Nobody noticed for two billion years. Then in 1972 a French chemist saw something off in a routine uranium shipment — a fraction slightly lower than it should be. Not by much. Just by a hair. Most people would have shrugged. He didn’t. He followed it. Francis Perrin Oklo discovery

​Lesson one, free of charge: the universe rewards people who refuse to round.

But here’s the part that should stop you.

When Oklo ran, it left fingerprints in the rock — specific atoms in specific ratios. Those fingerprints only look the way they do if the laws of physics two billion years ago were essentially identical to the laws of physics today. Not close. Not similar. Identical to within a few parts in ten million.

The strength of electricity. The way atoms bind. The reason chemistry works at all. All of it, locked in, before continental drift had stopped.

I’ve spent a career building instruments to measure things that don’t exist yet. Oklo humbles me. Nature built the instrument first, ran the experiment for 150,000 years, and waited two billion years for a chemist to check his math.

The reactor stopped two billion years ago. The impact hasn’t.

Until next time, have a M.A.G.I.C. Week,

Brian

Read more