I, too, can't be replicated...

It is very funny to see that authors who in 2019 argued that "Preregistration is redundant, at best", are now co-authoring papers where their collaborators want to preregister all studies. I guess they weren't even able to even convince their direct co-authors!

@nntaleb Probably should have included some other empires, like Arab and Persian? I'm sure there were more empires that occupied these crossroads, just don't know their names.

@nntaleb "City-states interested in commerce" is my impression of what was historically the norm for the entire eastern Mediterranean (including Palestine). Exceptions were when these city-states were part of a larger empire (Roman, Byzantine, Ottoman, European, etc.)

A central point is that Phoenician city states were not interested in war and domination via land control; with the exception of Carthage, they just wanted commerce.

@nntaleb A fair reading of Trump's remarks is, no, he didn't really really think Japan was an ally in 1941, but yes, he did really think they might leak his plans to Iran! Calligulino, yer welcome for the attention to this matter

So Calligulino really, really thought that Japan was an ally in 1941?

@thomaschattwill It may take decades, or forever, to restore this squandered reputation.

@thomaschattwill I grant your caveat but...the world now knows that we are only one (or two) bad elections away from the next unreliable, shoot from the hip President who can't be sufficiently contained or ameliorated by our other institutions (Congress, judiciary, administrative state, media)

The crucial caveat here is that it is The U.S. "under Donald Trump," no the U.S. per se

In any profession, 90% of people are clueless but work by situational imitation, narrow mimicry & semi-conscious role-playing. Except social "science" and journalism where it is 99% and 100%, respectively.

Classic Intellectual Yet Idiot (IYI)

@GAZAWOOD1 There's so much AI and other nonsense posted to social media you need to include info on the source so people know these are *real* examples of what they show

SHOCKING. Disgusting. Straight from hell‼️ A Muslim preacher recounts how a husband beat his pregnant wife so badly → SHE LOST THE BABY. Then he says: "Why did this woman annoy her husband? Don't you see how hard he works? DON'T UPSET YOUR HUSBAND!" 😡😡😡

@ScientificGems @HigherEd_UK University degrees are supposed to be avocational, not vocational. You work a job; you are *called* to a degree, or a profession. Like an artist who creates because they have to, not because they need to make a living

@HigherEd_UK University degrees ARE supposed to be vocational. But the best vocational degrees teach things that will be valuable throughout life, not just for the next 12 months.

This is why university degrees are not supposed to be vocational.

We're thrilled to announce the Ctrl-Z Award, a US$2,500 prize for researchers “who discover substantial errors in their published work and take meaningful steps to correct the scientific record." In @Nature today; more here: centerforscientificintegrity.org/2026/03/10/ann… nature.com/articles/d4158…

@nntaleb Maestro, you are correct we are expending million dollar munitions to destroy fake or low value targets, but the China live video you posted is fake, likely AI. Please do better.

If true, big. History is full of unexpected adaptations to aggression, like antibiotic resistance.

@anderssandberg All models are wrong, some are useful. Hoping for the later!

It will be exciting to see what works and doesn't work here (both tell us nontrivial things!) Plus, the commitment to emulated lab animal ethics.

@NickTimiraos Also, how do you calculate a 6 month average for part of 2026, when we're only in the 3rd month of 2026? How do you calculate a 3 month average for anything more recent than the first 7 days of 2026? Why do the 6 month and 3 month lines end at the same point?

While the strength in the January report was not revised away, the weakness in the February report drops the three-month average of monthly job growth to 6,000, and the six-month average sits at -1,000.

The economy shed 92,000 jobs in February, and the unemployment rate rose to 4.4%. Revisions subtracted 69,000 in December and January from previously reported job growth, leading to negative job growth in December.

@NickTimiraos As always, the recent data in this graph are rough, often biased estimates that are subject to *large* revision

@cyb3rops Possibly there could be a role for the junior staff to check and verify AI output? But not as many needed as compared to junior staff actually doing the work.

Ppl keep getting my AI takes wrong - mostly because they think they contradict each other. They don’t. Simplified, my view is this 1. AI will replace a significant part of the workforce, especially entry-level white-collar jobs. 2. AI output is not trustworthy. It will stay flawed while sounding completely convinced that it is right. These two things can both be true at the same time. AI does not need to be always right to replace people - because people aren’t always right either. It just needs to be good enough, while being 100 to 1000 times faster and cheaper. That’s the part many still don’t get.

@HowidyHamza This horrible attack is the first case, as far as I can tell, of settlers actually killing someone in the West Bank. All previous incidents I'm aware of have involved lesser violent crimes. If there have been more killings, where can I find more information about them? Thanks.

HORRIFIC | Israeli settler terrorists stormed the village of Susya in the West Bank using military uniforms, opening their fire on residents, killing 28 year old Amer Shnnaran and wounding two others.

@heynavtoor When we say a model hallucinates X% of the time, what we mean is, the model hallucinates 100% of the time, but 100-X% of the time the hallucinations are plausible or even useful.

🚨BREAKING: OpenAI published a paper proving that ChatGPT will always make things up. Not sometimes. Not until the next update. Always. They proved it with math. Even with perfect training data and unlimited computing power, AI models will still confidently tell you things that are completely false. This isn't a bug they're working on. It's baked into how these systems work at a fundamental level. And their own numbers are brutal. OpenAI's o1 reasoning model hallucinates 16% of the time. Their newer o3 model? 33%. Their newest o4-mini? 48%. Nearly half of what their most recent model tells you could be fabricated. The "smarter" models are actually getting worse at telling the truth. Here's why it can't be fixed. Language models work by predicting the next word based on probability. When they hit something uncertain, they don't pause. They don't flag it. They guess. And they guess with complete confidence, because that's exactly what they were trained to do. The researchers looked at the 10 biggest AI benchmarks used to measure how good these models are. 9 out of 10 give the same score for saying "I don't know" as for giving a completely wrong answer: zero points. The entire testing system literally punishes honesty and rewards guessing. So the AI learned the optimal strategy: always guess. Never admit uncertainty. Sound confident even when you're making it up. OpenAI's proposed fix? Have ChatGPT say "I don't know" when it's unsure. Their own math shows this would mean roughly 30% of your questions get no answer. Imagine asking ChatGPT something three times out of ten and getting "I'm not confident enough to respond." Users would leave overnight. So the fix exists, but it would kill the product. This isn't just OpenAI's problem. DeepMind and Tsinghua University independently reached the same conclusion. Three of the world's top AI labs, working separately, all agree: this is permanent. Every time ChatGPT gives you an answer, ask yourself: is this real, or is it just a confident guess?

@RexDouglass At no time since 1979 were the US and Iran at peace.

Should add the U.S. and Iran have been at war since at least 2003 and the invasion of Iraq.

Just so it's super clear. The U.S. has been fighting a hot proxy war with Russia through Ukraine for years.Russia and China are now fighting a hot proxy war with the U.S. through Iran. It's not World War 3, but you can see it from here.

@billcullen37 @PhysInHistory Hawking radiation is very weak, except for tiny black holes which quickly 'evaporate'. There are experiments under way to detect these in nature, or create them in the lab. #Experimental_observation" target="_blank" rel="nofollow noopener">en.wikipedia.org/wiki/Hawking_r…

@PhysInHistory Has Hawking radiation been observed experimentally?

One of the most amazing 'what ifs' in modern physics involves a blackboard and a cleaner. Physicist Alan Lightman recalls a day in the 1970s when he and two graduate students were explaining their work on spinning black holes to Feynman over lunch at Caltech. Feynman immediately recognized a classical analogue to a quantum process and, walking back to Lightman's tiny office, began working out the equations for what would later be known as Hawking radiation—the theory that black holes can emit energy. He derived the process of spontaneous emission from black holes right there on the blackboard. The group realized its potential importance, but by the time Lightman returned to his office the next morning to copy it down, the cleaning lady had wiped the board clean . While Stephen Hawking would famously publish the theory a year later, this anecdote shows how Feynman's unique insight had, for a brief moment, stumbled upon one of the most significant discoveries in theoretical physics.