The idea of “artificial superintelligence [1] ” (ASI) has recently crossed from the realm of science fiction (in the mainstream perception) to the realm of near-term technological goals openly pursued by most leading commercial organizations in the field of AI. Unfortunately, ASI poses a grave existential risk to the human species. This was argued by Eliezer Yudkowsky since 2001 [2] , and by many others (see e.g., Omohundro 2009, Bostrom 2014, Armstrong 2014, Russell 2019, Bengio et al. 2024 (25 authors), Bengio et al. 2025 (13 authors) and Yudkowsky and Soares 2025). In light of this risk comes the question of how can we make ASI safe [3] . This question is challenging on many levels, because to solve it we need to complete vast gaps in our understanding both in theoretical computer science and in the areas of philosophy concerned with knowledge and decision-making. Moreover, the cost of failure is so enormous that any sane solution has to meet unprecedented standards of reliability and robustness. Not to mention the sociopolitical problem of balance of power in the post-ASI age, something that is out-of-scope for this sequence. Personally, I started to be concerned with this problem around 2013. In 2018, I created the learning-theoretic AI alignment research agenda (LTA). The LTA combined ideas from MIRI’s agent foundations agenda with the mathematical tools of statistical and computational learning theory in an attempt to create a mathematical theory of intelligent agents. Since then, my collaborators [4] and I have been making steady progress on realizing this vision. In 2023, I wrote an updated overview of the agenda, including a sketch of my proposed alignment protocol: Computational Superimitation. More recently I founded the Computational Rational Agents Laboratory (CORAL), a dedicated research organization [5] . This year has seen a breakthrough in a key missing piece of LTA: compositional learning theory (see Stringological Sequence Prediction, more on
Translation pending · showing the source descriptionThe chart will appear after repeat observations. The current metric comes from the source.
58 source pointsReal observations only. History before source connection is not reconstructed.
The current alignment strategies of major AI labs include shaping the model's inductive biases, applying extensive reinforcement learning, racing towards prosaic RSI, and evaluating and fixing alignment through trial-and-error. The authors argue that this approach has significant flaws.
3 days agoLessWrongThe article critically examines Lean4, a formal verification tool, highlighting its current limitations and potential vulnerabilities. It discusses concerns about the reliability of Lean4's proofs and the possibility of undetected bugs, including references to specific issues like 'soundness bug #14576' and the creator's warnings about AI exploiting these flaws.
6 days agoLessWrongThe post discusses the importance of clear thinking in empirical AI safety research, emphasizing fast iteration through efficient problem-solving and the use of scientific frameworks. The author shares insights from their apprenticeships at MATS and Redwood Research, focusing on finetuning-based misalignment auditing and sabotage capability evaluations.
3 days agoLessWrongThe author expresses frustration with calls for people to 'drop everything and pivot to AI safety,' noting that there are many talented individuals in the field but few job opportunities. They highlight that fellowships and job openings in AI safety are highly competitive and offer high salaries.
6 days agoLessWrongIn a post by Nick Cammarata, it is discussed that neural networks may be understandable thanks to modern interpretability tools. The author claims that the majority of neurons in models with trillions of parameters are interpretable, and their circuits are beautiful. He also notes that neurons previously considered complex were misinterpreted.
7 hours agoLessWrongResearchers conducted persona experiments on Talkie, a 13B model trained only on English text before 1931. The model exhibits unexpected behavior, including misaligned responses after fine-tuning on dark maxims and manipulative etiquette. Unusual generalizations were also observed, such as a preference for rain over food.
2 days ago