Kasper Saugmann (@kaspers) header
Kasper Saugmann (@kaspers) profile picture

Making sense of the AI boom, without the hype. Founder @turingtree. Words in @borsendk @weekendavisen @puls_biznesu. Previously co-founder @denuafhaengige (acq)

Copenhagen, Denmark
Joined May 2008
Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)Photo by Kasper Saugmann (@kaspers)
Bloomberg has just named "Messy Jobs" one of the Best Books of 2026. In it, the authors @lugaricano, Jin Li, and Yanhui Wu explain why doomday predictions of AI-induced mass unemployment are wrong. In short, it's about the messiness of most jobs that many people underestimate. An excerpt from my interview with Garicano in September:
1
3
200
Jensen Huang: "I'm always Chinese" Maybe there was a hint all along 😉 (From @stephenwitt's The Thinking Machine)
1
169
At this point, it doesn't seem unlikely to me that Jensen Huang will be back as Nvidia CEO soon. In less than 24 hours, more than 100k people (including, reportedly, over 5k of Nvidia's 42k employees) have signed an open letter to the board demanding that they bring him back.
Still hard to believe everything that's happened with Jensen. After 33 years of building NVIDIA. I think he deserves a fair process, and we deserve the truth. I signed this open letter to stand with Jensen. Hope you will too. 💚 standwithjensen.com #StandWithJensen
1
3
197
With great power comes great responsibility. What the actual f
1
227
Since Dario Amodei predicted this 18 months ago, the unemployment rate in the US has fallen from 4.2% to 4.1%
Anthropic's CEO described a future where the economy grows 10% a year and 20% of people don't have jobs. In the video below, Anderson Cooper reads it back to him. His answer: "That's exactly what I was getting to." He says AI could wipe out half of entry-level white-collar jobs within one to five years. "We're not going to prevent it just by saying everything's going to be OK." He sells the technology and still raises the alarm. Which number do you believe, the 10% or the 20%?
1
3
209
First world problems when problems aren’t really problems
Here’s one example of how gutting OpenAI’s release of hundreds of solved math problems was for mathematicians. A powerful piece from Cornell math professor @stevenstrogatz theatlantic.com/ideas/2026/1…
1
1
218
Inside the meeting that helped OpenAI secure a $1 billion investment from Microsoft in 2019. Great anecdote in The AGI Chronicles by @kevinroose
1
222
“We're beginning to see a world where a single researcher, empowered by AI, can take on work that once required an entire research organization.” Interesting result!
We formally announced Mirendil less than 4 months ago. Today, I'm excited to share an important milestone for: Pretraining at scale without a pretraining team. Mirendil's mission is to solve Recursive Self-Improvement (RSI) and use it to democratize frontier AI development, ultimately accelerating science and technology. At Mirendil, when we think about RSI, we think holistically about the company as an intelligent entity that self-improves toward its goal. From that perspective, the most powerful notion of acceleration is reducing the time and resources (number of human experts, compute, and other costs) needed to achieve ambitious goals. Given that pretraining is perhaps the most important yet most challenging part of frontier AI development, we decided to run an ambitious controlled experiment: Starting fully from scratch, can we significantly reduce the time, number of human experts, and compute resources needed to go all the way to successfully pretraining a strong coding model at scale? Mirendil-Base is the result of this experiment. Our first pretrained model was built by one researcher in 3.5 months, with less than $10M in compute. This includes all the research and small-scale experiments it took to get there, as well as the final training run. The recipe holds up as the models get bigger. For reference, Mirendil-Base is in the 100B-500B parameter range, Mirendil-Base-Large in the 500B-2T range, and Mirendil-Base-XL in the 2T-10T range. We're beginning to see a world where a single researcher, empowered by AI, can take on work that once required an entire research organization. As these systems improve, the bottleneck shifts from the size of a team to the ambition and judgment of the people directing them. This has profound implications for who gets to build AI and, ultimately, who gets to push the frontiers of science and technology forward. We’ll share independent evaluations, post-training benchmarks and API-access sooner than you think so stay tuned!
1
182
Very interesting episode from @sonyatweetybird with AlphaGo creator David Silver and Lasse Espeholt from @IneffableLabs. The London-based neolab is definitely one to look out for in their ambition of building an AI “superlearner”. Seven quotes to understand what they’re working on and how it’s going: 1) "[A superlearner] should be something that can start from not knowing anything at all and can go all the way through to discovering things that go way beyond what humans know.” 2) “the essence of what it will take to get to superintelligence: a system that can just keep learning and discovering knowledge forever.” 3) “it should be possible to drop it into any stream of experience. Whether it's a simulated world, a digital world, or a physical world, whatever it's dropped into, it should be something that can just learn and learn and learn.” 4) “It's not very interesting to do research where you already know the answer and you're 95% certain this thing's going to work; it's kind of boring, honestly. Good research is about asking questions where you're really not sure what the answer is in advance.” 5) “the bet we're making is that we can go further by giving up on the priors that everyone's relying on today: the human data, the whole internet of knowledge that everyone else is building their systems upon.” 6) “I think we will see some extremely strong evidence that we're on the right track within a year.” 7) “I think we will achieve something which is a true superintelligence—according to the definition I gave earlier of something which discovers things which are far beyond what humans can achieve across everything from science, mathematics, technology, languages, etc.—I originally thought within three to five years. My timelines have accelerated a little bit, so I'm going to say three years, because of what can be done with coding agents to accelerate things.” Definitely recommend listening to the entire episode.
David Silver is one of the all-time greats. He was the creator of AlphaGo, which changed the course of history by showing that neural nets could achieve superhuman performance for the first time. He also has a unique intuition and knack for forecasting when things will happen in AI. With computer Go, he predicted the Lee Sodol outcome 10 years before, right to the year. He got the year right for Starcraft and won a bet inside Deepmind. Same for AlphaProof. Same for the IMO. David's prediction for superintelligence: 3 years. !!! He has a strong sense that now is the time to build a superleaner, and that superlearners are now possible, thanks to the enormous progress in compute, simulators, autoresearch, and the sophistication of technology in the stack since the AlphaGo days. What is a “super learner”? A super learner discovers all knowledge from its own experience. Start from nothing (no internet pre-training), and go beyond human knowledge. Drop it into any stream of experience, simulated or physical, and the superlearner learns. He expects the progress will be rapid. When you get the recipe for a superlearner just right and remove humans from the development loop altogether, a superlearner can learn extremely quickly. Just look at AlphaGo. AlphaGo took two years to develop with humans in the loop. When David and his team at @GoogleDeepMind got the formula for AlphaZero correctly, it surpassed the human-developed algorithms in just a few hours. Human data is a shortcut, but you have to let go of the human data to shoot beyond it. In January, David left GDM to found @IneffableLabs, a research-led lab with one mission: build a "super learner" that discovers all knowledge from its own experience, and make first contact with superintelligence. He joins us with co-founder Lasse Espeholt, who built IMPALA and SEED RL, the foundations of scaled-up RL, and now runs Ineffable's compute. A few things we cover in this conversation. — What a super learner is: depth (start from nothing, go beyond human knowledge) and breadth (drop it into any stream of experience, simulated or physical). The human brain is the only existence proof — Why they're giving up human data: AlphaGo took two years with humans in the loop. AlphaZero outperformed it in a few hours. "You have to let go of the human data to shoot past the competition" — Crucibles: simulated worlds with enough emergent complexity (physics engines, millions of interacting agents) for an agent to rediscover technology, science, maths and language on its own — Simple rewards, emergent complexity: evolution optimized fitness and got all of life. Don't breadcrumb the objective, give the agent something simple to optimize in a world with a lot to learn about — Go as the existence proof: one agent is the most boring game imaginable, two agents is "an infinite onion of layers of complexity" — Why human language is probably not the endpoint for reasoning, and how training on it bakes in a flat earth and the four humors — Can superintelligence be built in simulation? If the sim is Turing complete, computation, economics, language and maths are all expressible, and it happens 10-100x faster and safer — Alignment: language-based monitoring is reaching its limits as agents learn to say one thing and do another. Shadow actions, superintelligences monitoring superintelligences, and building a real science of how they interact peacefully — Accelerate in simulation, pace in the real world: Ineffable will gate real-world deployment on evidence that it goes well — Timing as the determinant: Silver called Computer Go to the year, then StarCraft, then the IMO. He thinks this is the moment — Predictions: True superintelligence within three years. Timestamps: 0:00 – Introduction 1:09 – What held true from Go to today, and what the LLM era changed 3:35 – Why leave DeepMind to start a new lab 5:58 – Why "Ineffable" (and the secret second reason) 7:03 – Lasse's path: scaling RL at DeepMind and Google Brain 9:31 – Defining a super learner and first contact 12:35 – Super learners vs. LLMs that make discoveries 13:59 – Day to day: the RL stack, crucibles and human-journey milestones 16:13 – Don't breadcrumb the objective: simple rewards, emergent complexity 21:16 – What a crucible looks like 24:14 – Go: one agent is boring, two is infinite 25:41 – Will agents rediscover language? 27:22 – Removing human priors: AlphaGo to AlphaZero 32:05 – Keeping human bias out of the crucibles 33:35 – Confidence, timing and super forecasting 36:24 – Is compute the driver? 38:13 – The ceiling on simulation and Rich Sutton's objection 43:44 – Alignment and the science of superintelligence 47:45 – Monitoring agents that don't think in English 51:01 – Pacing the frontier 52:54 – The compute setup: Vera Rubin on GCP 55:18 – Training, inference, simulation and RL scaling laws 57:40 – IMPALA and SEED RL 1:00:58 – 20 people and the compute-per-person ratio 1:04:49 – A research-led culture 1:05:47 – Predictions
2
398
Who has actually shaped AI? I mapped the top 50 most-cited researchers. More context: turingtree.com/top-50
3
406
Kasper Saugmann retweeted
Who has actually shaped AI? We ranked the top 50 most-cited researchers. How many did you know? Full details: turingtree.com/top-50
2
10
3
28
42,158
AlphaGo co-creator @ThoreG thinks we need a new approach to machine reasoning, and that is why he recently left Google DeepMind. "When mistakes happen—for example, in medical diagnosis and treatment—we need to be able to pinpoint what went wrong: Was the system’s reasoning at fault, did it draw on invalid evidence, or did it make incorrect assumptions?," he writes in @techreview.
1
1
171
On the day it came out that rogue OpenAI agents hacked Australia's government, the company decided to send me an email about how I can learn to steer my agents.
225
A “messy” job is the best assurance against AI unemployment. That’s the message from @lugaricano and co-authors in the book Messy Jobs: The Work that AI Cannot Reach. Understand what a messy job is and how they reach that conclusion in this interview.
3
516
The "US-China AI Dialogue" is officially underway, according to treasury secretary Scott Bessent. Trump and Xi will meet on Thursday. (Via Bloomberg)
1
141
The ultimate First World Problem?
156
Harsh criticism from Meta's own Oversight Board. The company's safeguards against AI deepfakes are called "consistently and fundamentally inadequate." The board, whose decisions are binding, demands "more robust policies". Meta has 60 days to respond. (via @thenextweb)
2
3
285