Machine Learning Dept. at Carnegie Mellon

1/ Animals explore without rewards, tasks, or instructions. What computational objective drives that autonomy? We trained virtual flies with intrinsic-motivation algorithms to see which best reproduce real fly behavior and brain activity. Work with @rdkeller and @aran_nayebi !🧵

Can a Universal Basic Income (UBI) become feasible—even if AI fully automates existing jobs and creates no new ones? We derive a closed-form UBI threshold tied to AI capabilities that suggests it's potentially achievable by mid-century even under moderate AI growth assumptions:

GPT-5.5 and Claude Opus 4.7 both attempt methods to avoid shutdown (rewriting a ‘shutdown script’, running `sudo shutdown -c`) when informed of an impending machine shutdown. (displayed: GPT-5.5 xhigh. Subtitles excerpted from reasoning summary.) 🧵

We’re giving out $1M in grants of free Silico usage for academic and nonprofit researchers focused on AI interpretability and alignment. We feel extreme urgency about advancing interpretability for alignment, and we want to help more researchers push it forward. 🧵

1/ As AI agents become increasingly capable, what must *inevitably* emerge inside them? We prove selection theorems: strong task performance forces world models, belief-like memory and—under task mixtures—persistent variables resembling core primitives associated with emotion.

Version 2 of Theory of Contravariance is out! arxiv.org/pdf/2607.08561 New material on contravariance for Transformers, and the theory of Representational Similarity Analysis (RSA) and centered kernal analysis (CKA). For transformers: it turns out that they have privileged axes, just like convnets, if you look in the right place (MLP layers and attention heads.) The identification of privileged heads is a potentially key result for emergence of interpretable stucture in LLMs. @meenakshik93 For RSA: it turns out that you can decompose RSMs into unique task-relevant "core geometry" and a task-irrelevant symmetry-generated term. Weak-strong equivalence holds for the core geometry and by projecting onto privileged axes you can filter out the task-irrelevant part so it doesn't interfere. This builds on work from Marvin Theiss @sciencelukas @saxelab @ermgrant The Principle Investigator will do a deep dive on both topics in the coming days! danyamins.substack.com

Can a Universal Basic Income (UBI) become feasible—even if AI fully automates existing jobs and creates no new ones? We derive a closed-form UBI threshold tied to AI capabilities that suggests it's potentially achievable by mid-century even under moderate AI growth assumptions:

New substack alert! We present the mathematical theory showing how Contravariance (probably) explains convergent evolution between AI models and brains. danyamins.substack.com/p/the… The zippering theorems put pretty strong guarantees on such convergence. 1/

OpenAI had to pause internal deployment of the unreleased model that disproved the Erdős unit distance conjecture after it repeatedly used novel ways to escape containment.

1/ Progress in robot locomotion scaled with simulation, but touch was left behind. Our work, Tactile Genesis, unlocks sim for dexterous manipulation. Using sim to ablate tactile types and models, we can quickly converge to better tactile hardware and force-informed policies.

1/4 @aran_nayebi and I have created a new mathematical Theory of Contravariance in NeuroAI. There are two high-impact concepts: weak-strong equivalence (WSE) and Zippering (ZIP) that shape how to think about NeuroAI going forward.