--- schema: 1 type: video title: >- World Models Need Causality, Not Pretty Pixels — Christopher Manning, Moonlake AI description: >- In 2007, Google already had a language model trained on 2 trillion tokens, the same order of magnitude as today's frontier models. Christopher Manning, Stanford professor, former director of the Stanford AI Lab and now at Moonlake AI, uses that as one stop in a tour of AI history. It runs from Dartmouth in 1956 and Shakey the robot to the rise of LLMs, and it ends at his argument for what comes next: embodied intelligence built on simulation. Manning argues that generative video like Genie 3 simulates observations but has no semantics underneath, so it can't support planning. Moonlake takes a single photo or short video and builds an action-conditioned world in code, with objects you can pick up, open and move. It even researches objects on the web to fill in what the camera can't see, like the tea bags inside a closed box. A loop inspired by Claude Code compares renders against reality to shrink the sim-to-real gap, with the goal of replacing 10,000 hours of teleoperation with simulation. The talk ends with Q&A on gaming, ontologies, physics and discovery in latent space. Speaker info: X/Twitter: @chrmanning (https://x.com/chrmanning) LinkedIn: https://www.linkedin.com/in/christopher-manning-011575 Website: https://nlp.stanford.edu/~manning/ Related links: Moonlake AI: https://moonlake.ai Timestamps: 0:00 Intro 1:02 Story time: a history of AI 2:02 Dartmouth 1956 and cybernetics 3:46 The first machine translation demo (1954) 5:41 The Stanford AI Lab, the Stanford Cart and Shakey 8:11 Why early AI happened at Stanford, not Berkeley 11:01 Language models, from Markov to Shannon to IBM 14:10 How we got to large language models 17:40 Why embodied AGI is the North Star 19:30 Why simulation: the 10,000-hour teleop problem 20:50 Shakey's world model and the model in your head 22:30 Action-conditioned world models 23:50 Genie 3 and the problem with pretty pixels 25:19 From a photo to a world you can act in 29:04 Filling in what the camera can't see 30:54 Building world models in code 32:54 Software eats the physical world 35:53 10,000 hours of simulation for free 38:08 Q&A: gaming, ontologies, physics and sim-to-real --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/4Gqg0HVe-AY-youtube-f87b469378d9d11c24d3 Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney --- canonical: >- https://liquidgold.cc/channel/aidotengineer/explore/video/world-models-need-causality-not-pretty-pixels-chris-00933ec26624 markdown: >- https://liquidgold.cc/channel/aidotengineer/video/world-models-need-causality-not-pretty-pixels-chris-00933ec26624.md llms: null parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-24T03:24:42.536Z' language: null children: [] videoId: 4Gqg0HVe-AY durationSeconds: 3096 publishedAt: '2026-09-24T13:30:23.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 12 endSeconds: 24 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=12' heldReason: rights - ordinal: 1 text: null startSeconds: 124 endSeconds: 138 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=124' heldReason: rights - ordinal: 2 text: null startSeconds: 138 endSeconds: 234 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=138' heldReason: rights - ordinal: 3 text: null startSeconds: 234 endSeconds: 246 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=234' heldReason: rights - ordinal: 4 text: null startSeconds: 359 endSeconds: 499 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=359' heldReason: rights - ordinal: 5 text: null startSeconds: 683 endSeconds: 856 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=683' heldReason: rights - ordinal: 6 text: null startSeconds: 866 endSeconds: 1070 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=866' heldReason: rights - ordinal: 7 text: null startSeconds: 1082 endSeconds: 1172 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1082' heldReason: rights - ordinal: 8 text: null startSeconds: 1184 endSeconds: 1255 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1184' heldReason: rights - ordinal: 9 text: null startSeconds: 1265 endSeconds: 1356 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1265' heldReason: rights - ordinal: 10 text: null startSeconds: 1368 endSeconds: 1439 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1368' heldReason: rights - ordinal: 11 text: null startSeconds: 1450 endSeconds: 1523 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1450' heldReason: rights - ordinal: 12 text: null startSeconds: 1533 endSeconds: 1749 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1533' heldReason: rights - ordinal: 13 text: null startSeconds: 1759 endSeconds: 1859 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1759' heldReason: rights - ordinal: 14 text: null startSeconds: 1870 endSeconds: 1979 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1870' heldReason: rights - ordinal: 15 text: null startSeconds: 1870 endSeconds: 1979 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1870' heldReason: rights - ordinal: 16 text: null startSeconds: 1992 endSeconds: 2153 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=1992' heldReason: rights - ordinal: 17 text: null startSeconds: 2167 endSeconds: 2289 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2167' heldReason: rights - ordinal: 18 text: null startSeconds: 2300 endSeconds: 2395 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2300' heldReason: rights - ordinal: 19 text: null startSeconds: 2407 endSeconds: 2534 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2407' heldReason: rights - ordinal: 20 text: null startSeconds: 2545 endSeconds: 2666 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2545' heldReason: rights - ordinal: 21 text: null startSeconds: 2677 endSeconds: 2783 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2677' heldReason: rights - ordinal: 22 text: null startSeconds: 2794 endSeconds: 2910 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2794' heldReason: rights - ordinal: 23 text: null startSeconds: 2910 endSeconds: 2920 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=2910' heldReason: rights - ordinal: 24 text: null startSeconds: 3015 endSeconds: 3096 watch: 'https://liquidgold.cc/watch/4Gqg0HVe-AY?t=3015' heldReason: rights --- # World Models Need Causality, Not Pretty Pixels — Christopher Manning, Moonlake AI In 2007, Google already had a language model trained on 2 trillion tokens, the same order of magnitude as today's frontier models. Christopher Manning, Stanford professor, former director of the Stanford AI Lab and now at Moonlake AI, uses that as one stop in a tour of AI history. It runs from Dartmouth in 1956 and Shakey the robot to the rise of LLMs, and it ends at his argument for what comes next: embodied intelligence built on simulation. Manning argues that generative video like Genie 3 simulates observations but has no semantics underneath, so it can't support planning. Moonlake takes a single photo or short video and builds an action-conditioned world in code, with objects you can pick up, open and move. It even researches objects on the web to fill in what the camera can't see, like the tea bags inside a closed box. A loop inspired by Claude Code compares renders against reality to shrink the sim-to-real gap, with the goal of replacing 10,000 hours of teleoperation with simulation. The talk ends with Q\&A on gaming, ontologies, physics and discovery in latent space. Speaker info: X/Twitter: @chrmanning (https://x.com/chrmanning) LinkedIn: https://www.linkedin.com/in/christopher-manning-011575 Website: https://nlp.stanford.edu/~manning/ Related links: Moonlake AI: https://moonlake.ai Timestamps: 0:00 Intro 1:02 Story time: a history of AI 2:02 Dartmouth 1956 and cybernetics 3:46 The first machine translation demo (1954) 5:41 The Stanford AI Lab, the Stanford Cart and Shakey 8:11 Why early AI happened at Stanford, not Berkeley 11:01 Language models, from Markov to Shannon to IBM 14:10 How we got to large language models 17:40 Why embodied AGI is the North Star 19:30 Why simulation: the 10,000-hour teleop problem 20:50 Shakey's world model and the model in your head 22:30 Action-conditioned world models 23:50 Genie 3 and the problem with pretty pixels 25:19 From a photo to a world you can act in 29:04 Filling in what the camera can't see 30:54 Building world models in code 32:54 Software eats the physical world 35:53 10,000 hours of simulation for free 38:08 Q\&A: gaming, ontologies, physics and sim-to-real --- Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page: https://ai.engineer/talks/4Gqg0HVe-AY-youtube-f87b469378d9d11c24d3 Join our upcoming AI Engineer conferences (see https://ai.engineer landing page): - AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers - AIE CODE in SF, November 10–12 — the top AI Coding conference returns - AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney --- Canonical: https://liquidgold.cc/channel/aidotengineer/explore/video/world-models-need-causality-not-pretty-pixels-chris-00933ec26624 · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md) > Search chunk text is held until public-source provenance is verified. ## Document chunks > Missing: no stored document chunks are available.