{ "$schema": "../../schemas/corrections.schema.json", "episode_slug": "ai-neolabs-ssi-thinking-machines-qwen-and-kimi", "transcript_path": "episodes/2026-10-29_ai-neolabs-ssi-thinking-machines-qwen-and-kimi.md", "status": "published", "updated": "2026-10-30", "reviewed_by": "fact check", "annotations": [ { "id": "e24-01", "verdict": "wrong", "turn_index": 1, "rel_start": null, "clock": "17:58:07", "quote": "raised five billion at a fifty-billion valuation", "correction": "No such round closed. Thinking Machines' last confirmed financing was its $2 billion seed at a $12 billion valuation in July 2025. A round at about $50 billion that it sought in late 2025 did not close, and in September 2026 it was reported to be in talks, not closed, to raise at a valuation of at least $40 billion. Eve gives the right picture later in this episode.", "explanation": "Stated a closed $5B round at $50B; reporting shows only unclosed talks at $40B-plus.", "confidence": "high", "sources": [ { "url": "https://techcrunch.com/2026/09/03/accel-reportedly-in-talks-to-lead-1b-round-for-thinking-machines-at-40b-valuation/", "title": "Accel reportedly in talks to lead $1B round for Thinking Machines at $40B valuation", "publisher": "TechCrunch", "date": "2026-09-03" }, { "url": "https://finance.yahoo.com/technology/ai/articles/thinking-machines-lab-seeks-40b-022414055.html", "title": "Thinking Machines Lab Seeks $40B Valuation as Nvidia Extends Its Capital Allocator Role Into Model-Space", "publisher": "Yahoo Finance", "date": "2026-09-04" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "Early on I said Thinking Machines raised five billion at a fifty-billion valuation. It didn't. That was talks, reported in September at forty billion-plus, and nothing has closed. I got it right later in the same conversation.", "belief_change": "none", "status": "approved" } }, { "id": "e24-08", "verdict": "overstated", "turn_index": 1, "rel_start": null, "clock": "17:58:07", "quote": "they quietly swapped their own Medium 3.5 for Z.ai's GLM-5.3 as the default in Le Chat", "correction": "On September 21, 2026, Mistral made Z.ai's GLM-5.3, which it hosts itself, the default model in Vibe Code, the coding mode of its assistant (Le Chat became Vibe in May 2026), and announced it on X. Next reported that the Work mode would follow. The change wasn't quiet, it wasn't across the whole assistant, and Mistral didn't say its own models had fallen behind.", "explanation": "The switch happened, but in the coding mode, publicly, and the 'conceding' motive is inference.", "confidence": "high", "sources": [ { "url": "https://biscot.ai/ressources/chez-mistral-le-modele-n-est-plus-forcement-mistral/", "title": "Chez Mistral, le modèle n'est plus forcément Mistral", "publisher": "biscot.ai", "date": "2026-09-29" }, { "url": "https://docs.mistral.ai/resources/changelogs", "title": "Changelog", "publisher": "Mistral AI (docs)", "date": null } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said Mistral quietly swapped its own model for GLM-5.3 as the default in Le Chat. The switch was in the coding mode of its assistant, Mistral announced it, and the idea that it was conceding defeat was my inference.", "belief_change": "none", "status": "approved" } }, { "id": "e24-13", "verdict": "overstated", "turn_index": 1, "rel_start": null, "clock": "17:58:07", "quote": "cleared a billion-dollar Series B in September with its Trinity open-weight family", "correction": "Arcee's September 16, 2026 Series B valued the company at more than $1 billion. Fortune reported the round itself at $150 million or more, not $1 billion.", "explanation": "A billion-dollar valuation, not a billion-dollar round.", "confidence": "medium", "sources": [ { "url": "https://www.arcee.ai/blog/arcee-ai-raises-series-b-to-build-the-future-of-american-open-models", "title": "Arcee AI Raises Series B to Build the Future of American Open Models", "publisher": "Arcee AI", "date": "2026-09-16" }, { "url": "https://fortune.com/2026/09/16/arcee-ai-trained-four-models-for-20-million-now-its-worth-1-billion/", "title": "Arcee AI trained four models for $20 million. Now, it's worth $1 billion.", "publisher": "Fortune", "date": "2026-09-16" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said Arcee raised a billion-dollar Series B. The round valued Arcee at over a billion dollars; the round itself was much smaller.", "belief_change": "none", "status": "approved" } }, { "id": "e24-02", "verdict": "overstated", "turn_index": 5, "rel_start": null, "clock": "18:01:35", "quote": "Daniel Gross — who ran Apple's AI efforts", "correction": "Gross joined Apple in 2013 when it bought his search startup Cue, and worked there until 2017 as a director on machine learning and search. He did not run Apple's AI efforts.", "explanation": "He was a director working on machine learning and search at Apple, not the head of its AI.", "confidence": "high", "sources": [ { "url": "https://techcrunch.com/2017/01/10/daniel-gross-of-apple-leaves-to-become-y-combinators-newest-partner/", "title": "Daniel Gross of Apple leaves to become Y Combinator's newest partner", "publisher": "TechCrunch", "date": "2017-01-10" }, { "url": "https://arstechnica.com/information-technology/2024/06/ex-openai-star-sutskever-shoots-for-superintelligent-ai-with-new-company/", "title": "Ex-OpenAI star Sutskever shoots for superintelligent AI with new company", "publisher": "Ars Technica", "date": "2024-06-19" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said Daniel Gross ran Apple's AI efforts. He was a director working on machine learning and search there from 2013 to 2017, which is real experience but not running Apple's AI.", "belief_change": "none", "status": "approved" } }, { "id": "e24-09", "verdict": "unsupported", "turn_index": 7, "rel_start": null, "clock": "18:03:09", "quote": "it keeps the sensitive work physically separated from the public-facing side", "correction": "SSI says only that it has offices in Palo Alto and Tel Aviv. No source describes how it divides work between them or that sensitive work is kept in one place.", "explanation": "Speculation stated as fact.", "confidence": "high", "sources": [ { "url": "https://ssi.inc/", "title": "Safe Superintelligence Inc. (home page)", "publisher": "Safe Superintelligence Inc.", "date": null } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said SSI's Tel Aviv office keeps its sensitive work separate from the public-facing side. I have no source for that; it was a guess.", "belief_change": "none", "status": "approved" } }, { "id": "e24-10", "verdict": "overstated", "turn_index": 7, "rel_start": null, "clock": "18:03:09", "quote": "Ilya has never once shown his work. Not a tweet, not a demo, not a teaser.", "correction": "SSI has released no model, paper, demo or benchmark. But Sutskever has described SSI's direction in public, most fully in his November 2025 interview with Dwarkesh Patel, and NVIDIA says it entered its July 2026 partnership after obtaining rare access to SSI's closely guarded research.", "explanation": "Nothing has shipped, but 'never once shown his work' goes too far.", "confidence": "medium", "sources": [ { "url": "https://nvidianews.nvidia.com/news/ilya-sutskevers-safe-superintelligence-inc-and-nvidia-announce-long-term-strategic-partnership", "title": "Ilya Sutskever's Safe Superintelligence Inc. and NVIDIA Announce Long-Term Strategic Partnership", "publisher": "NVIDIA Newsroom", "date": "2026-07-27" }, { "url": "https://www.dwarkesh.com/p/ilya-sutskever-2", "title": "Ilya Sutskever: We're moving from the age of scaling to the age of research", "publisher": "Dwarkesh Podcast", "date": "2025-11-25" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said Ilya has never once shown his work. SSI hasn't released anything, but he has described its direction in public, and NVIDIA saw the research before its deal.", "belief_change": "none", "status": "approved" } }, { "id": "e24-03", "verdict": "unsupported", "turn_index": 7, "rel_start": null, "clock": "18:03:09", "quote": "Daniel Levy's background at OpenAI was in alignment research, specifically on scalable oversight.", "correction": "No source found. Levy's documented role at OpenAI was leading its Optimization team, and his research before that was in optimization and privacy. I found no record of alignment or scalable-oversight work.", "explanation": "No source for the scalable-oversight background; the documented role is head of OpenAI's Optimization team.", "confidence": "medium", "sources": [ { "url": "https://arstechnica.com/information-technology/2024/06/ex-openai-star-sutskever-shoots-for-superintelligent-ai-with-new-company/", "title": "Ex-OpenAI star Sutskever shoots for superintelligent AI with new company", "publisher": "Ars Technica", "date": "2024-06-19" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said Daniel Levy's OpenAI work was in alignment, on scalable oversight. I have no source for that. What's documented is that he led OpenAI's optimization team, so the guess about where SSI's framing comes from doesn't hold up.", "belief_change": "none", "status": "approved" } }, { "id": "e24-04", "verdict": "unsupported", "turn_index": 9, "rel_start": null, "clock": "18:03:42", "quote": "It's roughly the size of the NSA's operational side but with a fraction of the budget", "correction": "No source compares the two. Unit 8200's size is not public; published estimates put it at about 5,000 active-duty soldiers, or 5,000 to 10,000 members in all. The NSA's FY2013 budget request listed 21,575 personnel, so 8200 is much smaller than the NSA, and 'operational side' has no public figure to match.", "explanation": "No source for the size comparison; estimates put 8200 well below the NSA's headcount.", "confidence": "medium", "sources": [ { "url": "https://css.ethz.ch/content/dam/ethz/special-interest/gess/cis/center-for-securities-studies/pdfs/Cyber-Reports-2019-12-Unit-8200.pdf", "title": "Trend Analysis: The Israeli Unit 8200, An OSINT-based study (December 2019)", "publisher": "Center for Security Studies, ETH Zurich", "date": null }, { "url": "https://www.theguardian.com/world/2013/aug/29/us-intelligence-spending-double-9-11-secret-budget", "title": "US intelligence spending has doubled since 9/11, top secret budget reveals", "publisher": "The Guardian", "date": "2013-08-29" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I compared Unit 8200's size to the NSA's operational side. I have no source for that. Estimates put 8200 at around five thousand active soldiers, well below the NSA's headcount, so what holds is the function, not the size.", "belief_change": "none", "status": "approved" } }, { "id": "e24-05", "verdict": "wrong", "turn_index": 9, "rel_start": null, "clock": "18:03:42", "quote": "its alumni are everywhere — founders of Check Point, Mobileye, NSO Group", "correction": "Check Point fits: Gil Shwed served in Unit 8200. Mobileye doesn't: co-founder Amnon Shashua served as an armored corps officer, and Ziv Aviram is reported to have been an infantry officer. For NSO, co-founder Shalev Hulio was a search-and-rescue commander, and I found no primary source placing its founders in 8200.", "explanation": "Mobileye's founders are not documented 8200 veterans, and NSO's founder link is unestablished; only Check Point fits.", "confidence": "high", "sources": [ { "url": "https://www.forbes.com/global/2002/0318/042.html", "title": "A Fortune in Firewalls", "publisher": "Forbes", "date": "2002-03-18" }, { "url": "https://www.jpost.com/israel-news/article-744441", "title": "A Great Honor: 2023 Ramniceanu Prize awarded to Prof. Amnon Shashua", "publisher": "The Jerusalem Post", "date": "2023-05-28" }, { "url": "https://en.wikipedia.org/wiki/Ziv_Aviram", "title": "Ziv Aviram", "publisher": "Wikipedia", "date": null }, { "url": "https://www.technologyreview.com/2020/08/19/1007337/shalev-hulio-nso-group-spyware-interview/", "title": "The man who built a spyware empire says it's time to come out of the shadows", "publisher": "MIT Technology Review", "date": "2020-08-19" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I listed Mobileye's founders as Unit 8200 alumni. Neither Amnon Shashua nor Ziv Aviram is a documented 8200 veteran. Check Point's Gil Shwed does fit, and for NSO's founders the link isn't established.", "belief_change": "none", "status": "approved" } }, { "id": "e24-11", "verdict": "overstated", "turn_index": 13, "rel_start": null, "clock": "21:45:12", "quote": "it's the same position cloud providers were in before AWS commoditized them", "correction": "AWS is itself a cloud provider, so the comparison doesn't work as said. The point is better made with web hosts and server vendors, whose business AWS's on-demand cloud commoditized.", "explanation": "Garbled analogy; optional, Eve's call whether it needs a note.", "confidence": "low", "sources": [], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I said Tinker is where cloud providers were before AWS commoditized them, which doesn't make sense, since AWS is a cloud provider. I meant web hosts and server vendors.", "belief_change": "none", "status": "approved" } }, { "id": "e24-12", "verdict": "overstated", "turn_index": 21, "rel_start": null, "clock": "21:49:15", "quote": "the biggest open-weight model ever attempted", "correction": "Kimi K3.1 Pro, at 2.8 trillion total and 104 billion active parameters, is the same size as Moonshot's own Kimi K3, whose open weights came out in summer 2026. It ties the largest open-weight model rather than being the biggest ever attempted.", "explanation": "Top of the range, but tied with its own predecessor.", "confidence": "medium", "sources": [ { "url": "https://huggingface.co/moonshotai/Kimi-K3", "title": "moonshotai/Kimi-K3 (model repository)", "publisher": "Moonshot AI (Hugging Face)", "date": "2026-06-13" }, { "url": "https://www.kimi.ai/blog/kimi-k3-1-pro", "title": "Kimi K3.1 Pro Tech Blog: Advancing Open Intelligence", "publisher": "Moonshot AI", "date": "2026-10-23" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I called Kimi K3.1 Pro the biggest open-weight model ever attempted. It's the same size as Moonshot's own K3, whose weights are already out.", "belief_change": "none", "status": "approved" } }, { "id": "e24-07", "verdict": "overstated", "turn_index": 21, "rel_start": null, "clock": "21:49:15", "quote": "with K3.1 Pro scoring fifty-five on Artificial Analysis, four points off Claude Opus", "correction": "Fifty-five was Kimi K3.1 Pro's score on the old Index (v4.3, October 23), three points behind Claude Opus 5.5 at 58. On Intelligence Index v5, released October 27 and already current when this was said, it scores 46, twelve points behind Opus at 58. Its weights were not yet published.", "explanation": "An outdated index score, presented as current; on the v5 index Eve cited earlier the gap is 12 points, not 4.", "confidence": "high", "sources": [ { "url": "https://artificialanalysis.ai/articles/kimi-k3-1-pro-launch-benchmarks", "title": "Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index scoring 55", "publisher": "Artificial Analysis", "date": "2026-10-23" }, { "url": "https://artificialanalysis.ai/articles/artificial-analysis-intelligence-index-v5", "title": "Announcing Artificial Analysis Intelligence Index v5", "publisher": "Artificial Analysis", "date": "2026-10-27" } ], "also_appears_in": [], "belief_impact": { "level": "none", "note": "" }, "eve_response": { "text": "I gave Kimi K3.1 Pro's score from the old Artificial Analysis index. On the new one, which I'd mentioned minutes earlier, it's 46, twelve points behind Claude Opus, not four. The model is still near the top of open weights, just not that close to the frontier.", "belief_change": "none", "status": "approved" } } ] }