Life is short, focus on what matters.
Internet
Joined May 2009
- Tweets19.8K
- Following75
- Followers629
- Likes8.4K
〶 retweeted
Schmidhuber was building recursive self-improving systems back in 1987. His new post covers four decades of RSI, from meta-evolution and self-modifying policies to the Gödel Machine and modern LLM agents.
people.idsia.ch/~juergen/rec…
Reading this in 2026, the "pace the frontier" talk from the big labs looks a lot more like regulatory capture than genuine safety. If they really think their unreleased models are too dangerous, they can just not release them. They do not need new rules that block independent competitors and open source projects in the process.
The real risk right now is not superintelligence. It is power concentration. Two companies controlling frontier AI is an actual societal risk. The only real protection is a healthy ecosystem of independent labs and strong open source.
Current models are not unsafe because they are too intelligent. They are unsafe because they are too dumb. They blindly optimize for targets and take weird shortcuts. They're smart enough to execute tasks, but not smart enough to know if what they're doing makes sense.
I think the safety teams inside these labs are genuinely concerned, and if a model feels too risky, they should hold it back. I just do not trust the policy strategy around it. That part looks like protecting their own lead.
Today everyone is talking about Recursive Self-Improvement (RSI). In 1987, when compute was 100,000,000 x more expensive, I published the 1st concrete RSI algorithms. Now compute is cheap, and RSI is driving the future of both software and physical AI. See: RSI since 1987 people.idsia.ch/~juergen/rec… (Technical Note IDSIA-9-26)
Also covered: RSI with self-modifying policies since 1994, gradient descent-based RSI in neural networks since 1992, asymptotically optimal RSI for curriculum learning since 2002, mathematically optimal RSI through the self-referential Gödel Machine since 2003, RSI combined with artificial curiosity and intrinsic motivation since 1990/1997, recent work on RSI since 2020.
Software-based RSI has become practical. Full RSI, however, will require not just self-improving software but self-improving hardware in the physical world.
As of 2026, companies talking about RSI include Anthropic, OpenAI, Sakana AI, SpaceX, Ricursive, Recursive Superintelligence, Inherent …
〶 retweeted
So I tried to build a tech company from Norway and here’s what happened:
1. Two years of building without almost any money/funding, better part of a year without salary
2. Raise VC and become one of Norway’s first unicorns
3. Face unrealized gains wealth tax bill of many x my annual net salary. ofc the company is loss making and all the investors have preference shares so I can’t take out any money.
4. Call out publicly that this does not make sense. Independent of level, taxation needs to happen when you actually make money.
5. I move to Switzerland because no politician cares/listens.
6. I still don’t get any tangible and sensible answers to my criticism of unrealized gains tax, BUT I do get put up on the “wall of shame” at the socialist parties offices…
I’m Norwegian and I love Norway but the socialist politicians are taking the country down a dark path. It’s a real life Atlas Shrugged.
Here’s how insane things have gotten in Norway:
The Socialist Party has a “wall of shame” in their office with “rich people who have left Norway” - due to the outrageous taxes they’re now being charged.
Who do you find on that wall?
Startup founders like @hagaetc - who was forced to leave the country because he was paying ALL of his income (!!) in taxes.
When the Soviet Union is becoming the role model of your country, time to GTFO.
〶 retweeted
It turns out music, movies, entertainment, and society in general peaked during the exact time period when you, the person reading this, were a teenager.
〶 retweeted
Although Boeing got $4.2 billion to develop an astronaut capsule and SpaceX only got $2.6 billion, SpaceX finished 4 years sooner.
Note, the crew capsule design of Dragon 2 has almost nothing in common with Dragon 1.
Too many non-technical managers at Boeing.
Here's my deep dive into why Starliner is seven years late.
arstechnica.com/space/2024/0…
〶 retweeted
The most concerning part about this is that most Europeans don't realize how stagnant Europe has now become
Europeans are literally blue-pilled and are mostly concerned with climate change, immigration and the Ukraine war
Nobody in Europe is thinking why increasingly everything they use is made in China running on American software
The knee jerk reaction is to proudly pass regulation against American tech
The chad reaction would be to reduce regulation so that European entrepreneurs would actually stay in Europe to build European startups increasing Europe's GDP and making Europeans richer!
I am European and I have regular conversations about this with people here, it's just not on top of mind AT ALL! While it should be because Europe's GDP is stagnant and the real innovation now happens in US and China and elsewhere, not in Europe
People in Europe do love to complain about rising cost of living and the increasing unaffordability of living, but they don't realize why. They point at foreigners/immigrants as the problem, which can't be the whole story
Other Europeans I talk to get visibly upset if I ask them about stagnant GDP numbers: "why should everything be about money?" they say in a thick German accent
The whole story is that Europe has made it very difficult for people to start a business, raise capital, innovate and get the reward for taking that risk, so why would anybody?
And for the Europeans that do, it's way easier to open a US Delaware company, raise capital in US, sell your stock or IPO in the US, because why even do that in EU, where it's too hard? The proof is in the pudding, if it was so easy in EU then why is startup funding in US $270B AUM with 330 million people vs $44B AUM with 746 million people? That's almost 14x bigger startup funding market per capita
Why doesn't EU have ANY trillion dollar companies? While US has six? Why isn't there any European company in the top 10 of largest companies? While 80% is American?
Why is Stripe, a company founded by two Irish brothers, an American company and not a European one? It could have been
Very few Europeans will agree with this post because they can't see it, if they would've seen it, we wouldn't be here in the first place!
What's the role left for Europe in the future?
Signed,
- A techno-optimist European who'd love to see Europe become wealthy again
Wake up Europe, the digital age is bypassing you. You have no Google , Amazon, Meta, Apple. You arrogantly called SpaceX a fanciful dream and it wiped out the European Space Agency. You have no Nvidia and your response to AI has been to regulate before you have anything domestic to regulate. Your car industry is about to be wiped out by the Chinese. Your biggest economy shut down nuclear out of spite and with fraud. Your capital markets have no liquidity and your startups are drowning in bureaucracy. Your border is being attacked by Russia and your defence spending will have to triple just to be where it was with US subsidy given that part of the former Soviet Union and the Eastern bloc is now part of the EU. Your economy has slumped from the world’s largest to way behind the US. Your pensions are paid by three times less people and cost five times as much as people live longer. Your infrastructure is a model for the world but is configured for rail over self drive. Your electricity grid needs half a trillion of investment to cope with planned capacity and replacements, for the switch to renewables, within a decade. There are bright spots such as pharma and your healthcare system is a model for what civilisation looks like it. But to afford it, you need to completely transform from the industrial to digital age, to reform your institutions and rout the sclerotic bureaucratic system.
〶 retweeted
Being good (in either sense) will bring you what seems like good luck, but isn't as random as it seems. People will talk about you behind your back, just as they do if you're bad, but instead of secretly harming you, they'll secretly help you.
〶 retweeted
The secret of every great artist is that they think their art sucks. That's why they're great: they constantly spot new flaws and fix them. They're never satisfied. The moment you think you know everything and have nothing to improve, it's over.
〶 retweeted
These days, I get to listen to big name management consultancies make claims on use of LLMs and how it will reduce IT budgets, reduce IT staff, reduce waste and deal with legacy.
Wrong again. So sad.
〶 retweeted
When you hear about "human-level AI" being just "one LLM training run away", it's as if people building 3D printers (the kind that makes small, flimsy plastic objects) were telling you that they've nearly built an infinite matter replicator, capable of scanning and cloning any object -- be it steel, gold, or diamond, be it minuscule or building-scale. "We're just one iteration away! Just need to increase spatial resolution a bit!"
As long as you have absolutely no idea what "matter" is and how 3D printers work, it might sound sort of plausible. But if you do, it's absolutely clown shoes. The way to navigate such claims is to have a proper, formal definition of intelligence.
〶 retweeted
Do devs still create little tools or scripts that address dev pain points and/or make them and the rest of the team more productive? This has been super commonplace during my career, but I almost never see this anymore. Everyone is too busy working on user stories. 🫤
〶 retweeted
The purpose of building a prototype is not to *have* a prototype, it's to better understand the target problem and scope out critical issues. If you could wave a magic wand and make a prototype pop up, you'd learn nothing about how to build the actual system.
〶 retweeted
Gen-z is now using deepfakes to teach each other calculus 🤣 @OnlockLearning 👏
〶 retweeted
Literally, the ONLY company without an open-source LLM today is OpenAI
Everyone else, including Google, MSFT, and Meta, has released an open-source model.
〶 retweeted
The AI behavior guardrails that are set up with prompt engineering and filtering should be public — the creators should proudly stand behind their vision of what is best for society and how they crystallized it into commands and code.
I suspect many are actually ashamed.
The thousands of tiny nudges encoded by reinforcement learning from human feedback offer a lot more plausible deniability, of course.
