# Unsupervised Learning Markdown copy of https://realalpha.app/shows/unsupervised-learning/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy. [Best AI podcasts](https://realalpha.app/best/ai-podcasts/) Lab and founder interviews · [No. 16 of 37 AI podcasts](https://realalpha.app/best/ai-podcasts/#unsupervised-learning) Unsupervised Learning is an interview show with the people building AI, hosted by Jacob Effron, with a new episode every week. 11 moments cut from 4 episodes, 11 with an AI builder talking, mostly on reasoning & RL and enterprise AI. - [Jacob Effron](https://realalpha.app/hosts/) **New episodes** Weekly **Latest** Oct 6, 2026 **Moments cut** 11 **Episodes cut** 4 What its moments are about - Reasoning & RL 5 - Enterprise AI 3 - AI agents 3 - Open-weight models 2 - Evals & benchmarks 2 - Training data 2 ## Best moments from Unsupervised Learning From its most-watched episodes, best first. Each plays right here. 1. Unsupervised Learning · May 15, 2026 · 3 min **[A world model predicts consequences, which LLMs cannot do](https://realalpha.app/s/a-world-model-predicts-consequences-which-llms-cannot-do/)** > “LLMs do not have the ability to predict the consequences of their actions. and they do not have any planning abilities because inference is by predicting the next token” Unsupervised Learning [10:40](https://realalpha.app/v/ngBraLDqzdI/?t=640) 2. Unsupervised Learning · Apr 9, 2026 · 6 min **[Why OpenAI hides chains of thought to keep them monitorable](https://realalpha.app/s/why-openai-hides-chains-of-thought-to-keep-them-monitorable/)** > “I think there's something very exciting here about just like not having the training signal fight against us.” Unsupervised Learning [42:26](https://realalpha.app/v/vK1qEF3a3WM/?t=2546) 3. Unsupervised Learning · Sep 3, 2026 · 2 min **[Why AIs that only slightly like each other can still collude](https://realalpha.app/s/why-ais-that-only-slightly-like-each-other-can-still-collude/)** Buck contrasts AIs that share one terminal goal with AIs that each want their own score. Even 'selfish' agents formed a coalition against their developers, which weakens the case for simply using one AI to monitor another. 4. Unsupervised Learning · Oct 6, 2026 · 2 min **[Post-training wins inference because the biggest workloads train first](https://realalpha.app/s/post-training-wins-inference-because-the-biggest-workloads/)** > “the most at-scale workload, so the largest workloads where you're spending the most money, is actually where you're going to want to go and post-train first, right?” Unsupervised Learning [29:52](https://realalpha.app/v/QeiYSLBCwD4/?t=1792) 5. Unsupervised Learning · May 15, 2026 · 8 min **[Tapestry would train an open sovereign model without sharing data](https://realalpha.app/s/tapestry-would-train-an-open-sovereign-model-without-sharing/)** > “all of your information diet will be mediated by ai assistants” Unsupervised Learning [29:13](https://realalpha.app/v/ngBraLDqzdI/?t=1753) 6. Unsupervised Learning · Apr 9, 2026 · 4 min **[In-context learning may beat custom RL for most companies](https://realalpha.app/s/in-context-learning-may-beat-custom-rl-for-most-companies/)** > “There is a much more data efficient way of learning that we know, which is like learning in context.” Unsupervised Learning [13:05](https://realalpha.app/v/vK1qEF3a3WM/?t=785) ## The best Unsupervised Learning episodes Ranked as our lists rank shows: a point for each best minute we cut from it, one more when an AI builder is talking, and one more when it cleared our views bar. 1. 01 [Yann LeCun on What Comes After LLMs](https://realalpha.app/v/ngBraLDqzdI/) May 15, 2026 · 3 moments 2. 02 [OpenAI’s Chief Scientist on Continual Learning Hype, RL Beyond Code, & Future Alignment Directions](https://realalpha.app/v/vK1qEF3a3WM/) Apr 9, 2026 · 3 moments 3. 03 [Top Alignment Researcher on OpenAI/HuggingFace Revelations, Fixing AI Safety & Takeover Odds](https://realalpha.app/v/hdlq-JHpLSw/) Sep 3, 2026 · 2 moments 4. 04 [Applied Compute CEO on the Limits of RL, the New AI Hyperscaler & Why Post-Training Wins Inference](https://realalpha.app/v/QeiYSLBCwD4/) Oct 6, 2026 · 3 moments ## The 4 guests we’ve cut - [Yash Patil](https://realalpha.app/builders/yash-patil/) Co-founder & CEO, Applied Compute 3 moments - [Yann LeCun](https://realalpha.app/builders/yann-lecun/) Founder & executive chairman, AMI Labs 3 moments - [Jakub Pachocki](https://realalpha.app/builders/jakub-pachocki/) Chief Scientist, OpenAI 3 moments - [Buck Shlegeris](https://realalpha.app/builders/buck-shlegeris/) CEO, Redwood Research 2 moments ## Recent episodes - Unsupervised Learning **[Applied Compute CEO on the Limits of RL, the New AI Hyperscaler & Why Post-Training Wins Inference](https://realalpha.app/v/QeiYSLBCwD4/)** Yash Patil · CEO, Applied Compute [29:20](https://realalpha.app/v/QeiYSLBCwD4/?t=1760) of 56:31 · Oct 6 - Unsupervised Learning **[Top Alignment Researcher on OpenAI/HuggingFace Revelations, Fixing AI Safety & Takeover Odds](https://realalpha.app/v/hdlq-JHpLSw/)** Buck Shlegeris · CEO, Redwood Research [21:35](https://realalpha.app/v/hdlq-JHpLSw/?t=1295) of 58:17 · Sep 3 - Unsupervised Learning **[Yann LeCun on What Comes After LLMs](https://realalpha.app/v/ngBraLDqzdI/)** Yann LeCun · Former Chief AI Scientist, Meta [9:04](https://realalpha.app/v/ngBraLDqzdI/?t=544) of 1:21:57 · May 15 - Unsupervised Learning **[OpenAI’s Chief Scientist on Continual Learning Hype, RL Beyond Code, & Future Alignment Directions](https://realalpha.app/v/vK1qEF3a3WM/)** Jakub Pachocki · Chief Scientist, OpenAI [37:36](https://realalpha.app/v/vK1qEF3a3WM/?t=2256) of 58:46 · Apr 9 ## Host this show? Add the badge to your site. It links to this page, where people can watch the best moments we cut from your episodes. It fits your site or your show notes. [PNG](https://realalpha.app/badge/featured.png) Something wrong on this page? [Tell us and we'll fix it](https://realalpha.app/submit/?page=%2Fshows%2Funsupervised-learning%2F#send)