[00:00] A decade ago, most of us would have been stunned  by the idea that a computer program can write   an entire book, let alone a scientific paper.  But come September 2026 and the headlines read   [00:12] “Using AI, a professor wrote 200 papers  this year” and it doesn’t even surprise   anyone anymore. We have gotten so used to AI  so quickly that it’s easy to forget just how   much it’s outperformed expectations. At least  some of them. Today I have a quick rundown of   [00:28] these wrong predictions, and some thoughts  on what we can learn from this. Make sure   you stick around till the end. I've got a  great offer from today's sponsor, Outskill.  Last year, a group of AI researchers asked more  than 200 experts how likely it was that artificial   [00:47] intelligence would solve or substantially help  solve one of the Millennium Prize Problems by   the end of 2027. The median answer was 10%.  AI hit this benchmark in September 2026. [01:01] The same group of experts was asked how well  the best AI would perform during 2026 on an   extremely difficult programming test called  LiveCodeBench Pro. At the time, the best score   was 7.7%. The median expert prediction was that  this would rise to 14% by the end of 2026. By May,   [01:21] AI had reached 53.8%. So the actual result  was almost 4 times the experts’ prediction,   In April and May 2026, experts were asked about  something called the AI task-completion time   [01:37] horizon. Roughly speaking, this measures how  long a coding job would take a human expert   when an AI can still complete it reliably.  The experts predicted that by the end of 2026,   AI would reach about 3.4 hours.  While the survey was still running,   [01:52] a new model had already  reached 3 hours and 6 minutes. This isn’t a recent thing, it’s  been going on for a few years now. In autumn 2023, more than 2000 AI researchers  were asked how long various advances would   [02:10] take. They estimated it’d be 22 years until AI  could prove mathematical theorems publishable in   leading mathematics journals. Today’s reality  is that the maths and physics preprint server   [02:23] has announced they are being flooded with  papers that are, by admission of the author,   AI-written and that even the authors do not  understand. The new part isn’t that they don’t   understand their paper, the new part is that  they admit it. And the question is not any   [02:37] longer whether the papers are publishable,  but who is responsible for the content. In 2023 cybersecurity experts said that  AI wouldn’t be able to autonomously find   [02:49] and exploit vulnerabilities for years  to come. AI hit the mark the next year.   In 2022, experts estimated that AI  would reach gold level skills in the   international Mathematical Olympiad  by 2030. AI reached it in 2025. [03:02] So you see the pattern there, experts have  for many years underestimated AI progress.   My personal speculation for why this is is that  many of the so-called experts work in academia.   They’re used to things taking forever. They  don’t understand that market economies operate   [03:18] at a completely different pace, especially when  bolstered by hundreds of billions of dollars. But there’s a flipside. If you look at what  experts predicted would happen to jobs,   they were largely wrong. For example, one of  the jobs that was supposed to be automated   [03:34] first were truckers. This hasn’t even really  begun to happen. In 2020, the World Economic   Forum expected nearly half of work to be  automated by 2025. In a follow-up from 2025,   [03:46] they found that the actual rate was more like 22%.  And I tried to get ChatGPT to write this script,   but it still can’t tell relevant from  irrelevant information. It has, however,   mastered the important skill of sounding  very confident while wasting my time. [04:00] The most telling failure though is  Leopold Aschenbrenner’s prediction   of the impending intelligence explosion by  2027. This, if you remember was supposed to   happen by building a few 100 GigaWatt power  plants by drilling for a little more gas.   [04:17] And that in return was supposed to be done  by robots who were built by the AI run on   the gas-powered supercomputers that were  built by the robots. You see the problem. That the Silicon Valley bros  let this guy handle $30 billion   [04:32] tells you how far gone they are. Because really  the picture here isn’t so difficult to see. How   fast you can change reality depends dramatically  on how many physical changes you need to perform,   [04:44] and how many institutional hurdles stand in the  way. Maths, coding, and scientific publications   are a straight entry with near zero resistance.  But building a lab, building autonomous vehicles,   [04:56] building robots, changing workflows in  existing companies, building power plants,   mining resources, etc. That’s a completely  different thing. It will take decades. And from that we can also extrapolate  what’s going to happen in the near future.   [05:10] Rapid progress in anything that  can be done by software alone,   especially if that’s tasks which previously  were impossible, such as reconciling all   mathematics literature, which has now become  possible. There are many theoretical problems   [05:25] in physics and related disciplines that will be  up for grabs soon. Because they are solvable.   And the techniques for solving them are  already in the literature. Maybe physics   isn’t waiting for a new Einstein, it’s just  waiting for someone to read all the papers! [05:40] The most exciting thing is that this is not  exclusively something that AI companies can   do. Everyone can contribute. And that, I think, is  the future we’re headed into. The glorious phase   [05:52] in which everyone is a scientist, for a few years.  And then… Well, then we will either all die,   or never work again, depending on whom  you ask. I guess the lesson here is:   don’t listen to the experts. Follow  me for more uninformed opinions.. [06:07] Artificial intelligence is developing rapidly, and  OpenAI's GPT-6 Astra is closer to AGI than ever.   Are you keeping up? Are you sure? Today's sponsor,  Outskill, will bring you up to speed. Outskill's   [06:21] offering a GPT-6 Astra crash course. It's 3 hours  long, and it'll teach you how to build your own   Astra workflow hands-on. Outskill is an official  OpenAI partner, so their team has exclusive access   [06:35] before anyone else. They've tested it upside down  and had time to figure out how to get the most   out of it. In the workshop, you'll learn,  for example, how you can use Astra to find   jobs that are tailored to your skills, or how to  alert you of information concerning your stocks.   [06:51] When you sign up, you'll also get the Astra power  playbook with prompt formulas and workflows ready   for you to plug and play. The link to sign-up's  in the description, or you can scan the QR code on   [07:05] screen. I've signed up for it because it's an easy  way to stay up-to-date. So, go ahead, sign up for   this, and learn what less than 1% on this planet  know. Thanks for watching. See you tomorrow.