62 claims
| Verdict | Person | Kind | Said | Claim | Topic |
|---|---|---|---|---|---|
| Resolved false | Yann LeCun | Factual | 1 Oct 2026 | The OpenAI agents involved in the Hugging Face hacking incident were doing exactly what they had been asked to do. Yann LeCun · Factual · 1 Oct 2026 | AI safety and alignment |
| Resolved false | Bill Ackman | Factual | 1 Oct 2026 | Ackman estimated, with heavy hedging, that the average tenure of an S&P 500 CEO is probably about 3 to 4 years. Bill Ackman · Factual · 1 Oct 2026 | Economics and markets |
| Resolved false | Jensen Huang | Factual | 1 Jun 2026 | NVIDIA's token cost is the lowest in the world, lower than others by large factors and orders of magnitude, due to extreme co-design. Jensen Huang · Factual · 1 Jun 2026 | AI compute, hardware and data |
| Resolved false | Demis Hassabis | Promise | 20 May 2026 | Google DeepMind will release a Gemini 3.5 Pro model following Gemini 3.5 Flash. Demis Hassabis · Promise · 20 May 2026 | Their own work or organisation |
| Resolved false | Elon Musk | Promise | 28 Jan 2026 | Tesla hopes to debut the next-generation Roadster in April 2026. Elon Musk · Promise · 28 Jan 2026 | Their own work or organisation |
| Resolved false | Elon Musk | Promise | 28 Jan 2026 | Tesla will probably unveil Optimus 3 in a few months. Elon Musk · Promise · 28 Jan 2026 | Their own work or organisation |
| Resolved false | Dario Amodei | Prediction | 10 Mar 2025 | Within three to six months of March 2025, AI will be writing 90 percent of code. Dario Amodei · Prediction · 10 Mar 2025 | AI timelines and forecasting |
| Resolved false | Dario Amodei | Prediction | 10 Mar 2025 | Within twelve months of March 2025, AI may be writing essentially all of the code. Dario Amodei · Prediction · 10 Mar 2025 | AI timelines and forecasting |
| Resolved false | Dario Amodei | Prediction | 11 Nov 2024 | Amodei guessed that within about 10 months AI models would probably get pretty close on typical real-world programming tasks, reaching at least 90%. Dario Amodei · Prediction · 11 Nov 2024 | AI capabilities |
| Resolved false | Ray Kurzweil | Factual | 9 Jul 2024 | Moderna gathered several billion possibilities for what could cause COVID and used simulated trials with AI to narrow them. Ray Kurzweil · Factual · 9 Jul 2024 | Biosecurity |
| Resolved false | Nouriel Roubini | Prediction | 30 Jun 2024 | The RN (Le Pen's party) will win an absolute majority in the 2024 French parliamentary elections, with 60% probability as the baseline scenario. Nouriel Roubini · Prediction · 30 Jun 2024 | AI timelines and forecasting |
| Resolved false | Warren Buffett | Promise | 5 May 2024 | Poor Charlie's Almanack will be the only book at the Berkshire meeting bookstore next year. Warren Buffett · Promise · 5 May 2024 | Their own work or organisation |
| Resolved false | Elon Musk | Promise | 24 Jan 2024 | Musk said he thinks Tesla has a good chance of shipping some number of Optimus units in 2025, and he called Optimus a brand-new product. Elon Musk · Promise · 24 Jan 2024 | Their own work or organisation |
| Resolved false | Elon Musk | Promise | 24 Jan 2024 | Musk said Tesla's current schedule is to start production toward the end of 2025, sometime in the second half, while noting that he is often optimistic regarding time. Elon Musk · Promise · 24 Jan 2024 | Their own work or organisation |
| Resolved false | Rodney Brooks | Prediction | 1 Jan 2024 | After its first crewed test flight in April 2024, Boeing's Starliner is expected to fly with a crew once a year. Rodney Brooks · Prediction · 1 Jan 2024 | Technology |
| Resolved false | Jim Cramer | Prediction | 24 May 2023 | US politicians will cause investors to lose more money as the May 2023 debt ceiling negotiations drag on. Jim Cramer · Prediction · 24 May 2023 | Economics and markets |
| Resolved false | Jim Cramer | Prediction | 24 May 2023 | If the market follows the 2011 debt ceiling pattern, US stocks will fall about 12% from their level on 2023-05-24 to the bottom. Jim Cramer · Prediction · 24 May 2023 | Economics and markets |
| Resolved false | Eliezer Yudkowsky | Prediction | 6 Apr 2023 | The current AI developers will not try the alignment approach he described (filtering training data to nice examples and training on it). Eliezer Yudkowsky · Prediction · 6 Apr 2023 | AI safety and alignment |
| Resolved false | Jeremy Grantham | Prediction | Jan 2023 | The final low of this bear market might come well into 2024. Jeremy Grantham · Prediction · Jan 2023 | Economics and markets |
| Resolved false | Nouriel Roubini | Prediction | 6 Nov 2022 | The coming stagflationary debt crisis will be much worse than the 1970s and probably as bad as the 2008 Great Financial Crisis. Nouriel Roubini · Prediction · 6 Nov 2022 | Economics and markets |
| Resolved false | Nouriel Roubini | Prediction | 6 Nov 2022 | History suggests it will be nearly impossible for the Fed to avoid a hard landing in the US economy. Nouriel Roubini · Prediction · 6 Nov 2022 | Economics and markets |
| Resolved false | Nouriel Roubini | Prediction | 6 Nov 2022 | The coming US recession will not be short and shallow. Nouriel Roubini · Prediction · 6 Nov 2022 | Economics and markets |
| Resolved false | Dario Amodei | Prediction | 13 Oct 2022 | The situation where several companies can produce good reasoning models will rapidly cease as everyone moves further up the scaling curve. Dario Amodei · Prediction · 13 Oct 2022 | Economics and markets |
| Resolved false | Bill Ackman | Factual | 2 Aug 2022 | As of 2 August 2022, Visa had not responded publicly to the allegations about facilitating child trafficking on MindGeek/PornHub sites. Bill Ackman · Factual · 2 Aug 2022 | Other |
| Resolved false | GiveWell | Factual | 6 Apr 2022 | GiveWell is launching a contest for critiques of its cost-effectiveness analyses. GiveWell · Factual · 6 Apr 2022 | Science and research methods |
| Resolved false | Elon Musk | Guidance | 27 Jan 2022 | Tesla expects growth in 2022 over 2021 comfortably above 50%. Elon Musk · Guidance · 27 Jan 2022 | Their own work or organisation |
| Resolved false | Elon Musk | Guidance | 27 Jan 2022 | If able to respond to demand, Tesla's energy business might grow by about 200% to 300% in 2022 rather than about 50%. Elon Musk · Guidance · 27 Jan 2022 | Their own work or organisation |
| Resolved false | Elon Musk | Prediction | 27 Jan 2022 | Musk said he would be shocked if Tesla did not achieve full self-driving safer than a human in 2022. Elon Musk · Prediction · 27 Jan 2022 | AI capabilities |
| Resolved false | Gary Marcus | Prediction | 14 Dec 2021 | Gary Marcus bet that hybrid (neurosymbolic) AI models would pull ahead in 2022. Gary Marcus · Prediction · 14 Dec 2021 | AI timelines and forecasting |
| Resolved false | Elon Musk | Promise | 3 Mar 2021 | SpaceX will attempt a Starship launch on 4 March 2021, with the window opening at 9am. Elon Musk · Promise · 3 Mar 2021 | Their own work or organisation |
| Resolved false | Elon Musk | Prediction | 27 Jan 2021 | Tesla Semi will likely be the first vehicle to achieve full autonomy. Elon Musk · Prediction · 27 Jan 2021 | Their own work or organisation |
| Resolved false | Elon Musk | Prediction | 27 Jan 2021 | Musk said he was highly confident Tesla's cars would drive themselves with reliability in excess of a human in 2021. Elon Musk · Prediction · 27 Jan 2021 | AI capabilities |
| Resolved false | Elon Musk | Prediction | 27 Jan 2021 | Safe Level 5 full self-driving can be achieved using only the current Full Self-Driving computer version 1. Elon Musk · Prediction · 27 Jan 2021 | AI capabilities |
| Resolved false | Elon Musk | Guidance | 27 Jan 2021 | Tesla targets 100 gigawatt-hours of Tesla-produced cells in 2022. Elon Musk · Guidance · 27 Jan 2021 | Their own work or organisation |
| Resolved false | Ørsted | Prediction | Feb 2018 | Ørsted expects gross investments in 2018 of DKK 16-18 billion, covering offshore wind projects, the Asnæs Power Station biomass conversion and smart meter installation. Ørsted · Prediction · Feb 2018 | Their own work or organisation |
| Resolved false | Ørsted | Prediction | Feb 2018 | Ørsted expects 2018 EBITDA (business performance) excluding new partnership agreements of DKK 12-13 billion, compared with DKK 12.7 billion in 2017. Ørsted · Prediction · Feb 2018 | Their own work or organisation |
| Resolved false | Scott Alexander | Prediction | Jan 2018 | Scott Alexander gave an 80% probability that Bitcoin would be priced above $10,000 at the end of 2018. Scott Alexander · Prediction · Jan 2018 | Economics and markets |
| Resolved false | Scott Alexander | Prediction | Jan 2018 | Scott Alexander gave a 70% probability that the first Falcon Heavy launch would not be successful. Scott Alexander · Prediction · Jan 2018 | Technology |
| Resolved false | Scott Alexander | Prediction | 2018 | If global existential risks became a big part of 2018-2023, it would be because someone did something incredibly stupid or awful with infectious diseases. Scott Alexander · Prediction · 2018 | Other global risks |
| Resolved false | Nassim Taleb | Factual | 10 Jul 2017 | The United States is a federation, not a republic. Nassim Taleb · Factual · 10 Jul 2017 | Other |
| Resolved false | Elon Musk | Promise | 16 Jun 2017 | SpaceX plans to send a Dragon 2 propulsive lander to Mars in about two years, and probably another Dragon mission in 2020. Elon Musk · Promise · 16 Jun 2017 | Their own work or organisation |
| Resolved false | Ørsted | Promise | 2017 | DONG Energy will stop using coal altogether at its power plants from 2023. Ørsted · Promise · 2017 | Their own work or organisation |
| Resolved false | Ørsted | Prediction | 2017 | DONG Energy's REnescience waste treatment plant in Northwich is expected to be commissioned in the first half of 2017. Ørsted · Prediction · 2017 | Their own work or organisation |
| Resolved false | Elon Musk | Promise | 20 Jul 2016 | Tesla plans to create an integrated solar-roof-with-battery product and scale it throughout the world. Elon Musk · Promise · 20 Jul 2016 | Their own work or organisation |
| Resolved false | Scott Alexander | Prediction | 25 Jan 2016 | Scott Alexander gave an 80% probability that, if Trump won the Republican nomination, he would lose the 2016 general election. Scott Alexander · Prediction · 25 Jan 2016 | Other |
| Resolved false | Scott Alexander | Prediction | 25 Jan 2016 | Scott Alexander gave a 60% probability that ISIS would not continue to exist as a state entity by the end of 2016. Scott Alexander · Prediction · 25 Jan 2016 | Other global risks |
| Resolved false | Scott Alexander | Prediction | 25 Jan 2016 | Scott Alexander gave a 60% chance that Hillary Clinton would win the 2016 US Presidency. Scott Alexander · Prediction · 25 Jan 2016 | Other |
| Resolved false | Bill Ackman | Prediction | Jan 2016 | Pershing Square did not believe the events that hit Valeant in 2015 would permanently impair Valeant's intrinsic value. Bill Ackman · Prediction · Jan 2016 | Economics and markets |
| Resolved false | Jeremy Grantham | Prediction | Jan 2016 | Oil stocks should do well over the next five years, perhaps regaining much of their losses. Jeremy Grantham · Prediction · Jan 2016 | Economics and markets |
| Resolved false | Nate Silver | Prediction | 2015 | Donald Trump will not win the 2016 Republican presidential nomination. Nate Silver · Prediction · 2015 | Other |
| Resolved false | Nate Silver | Prediction | 2015 | In all likelihood, Trump could not win the Republican nomination with only a plurality of votes or delegates, because the Republican Party would unite against him. Nate Silver · Prediction · 2015 | Other |
| Resolved false | Marc Andreessen | Prediction | 20 Aug 2011 | Within 10 years of 2011, at least five billion people worldwide will own smartphones. Marc Andreessen · Prediction · 20 Aug 2011 | Technology |
| Resolved false | Vinod Khosla | Prediction | 23 Dec 2010 | Traditional lithium-ion bulk batteries will almost certainly not be around in 15 years. Vinod Khosla · Prediction · 23 Dec 2010 | Technology |
| Resolved false | Ray Dalio | Prediction | Jun 2010 | The next US recession will come sooner than the normal roughly five-year interval, because of the size of the debt. Ray Dalio · Prediction · Jun 2010 | Economics and markets |
| Resolved false | Jeremy Grantham | Prediction | Oct 2009 | GMO's seven-year forecast return for the S&P 500 is well below 3%, while the high-quality subset is still handsome. Jeremy Grantham · Prediction · Oct 2009 | Economics and markets |
| Resolved false | Jeremy Grantham | Prediction | Oct 2009 | The irony may well be that strong economic data in 2010 is likely to be accompanied by a weak stock market. Jeremy Grantham · Prediction · Oct 2009 | Economics and markets |
| Resolved false | Vinod Khosla | Prediction | 28 Sep 2007 | Khosla said he believed 50 million acres of US land could supply all US biofuel needs. Vinod Khosla · Prediction · 28 Sep 2007 | Technology |
| Resolved false | Jeremy Grantham | Prediction | Apr 2007 | GMO's 7-year forecast implies U.S. equities would return minus 1.4% real per year as P/Es and profit margins revert to long-term averages. Jeremy Grantham · Prediction · Apr 2007 | Their own work or organisation |
| Resolved false | Nouriel Roubini | Prediction | 25 Sep 2006 | New York City home prices will be 5 to 10 percent lower on average a year from September 2006. Nouriel Roubini · Prediction · 25 Sep 2006 | Economics and markets |
| Resolved false | Ray Kurzweil | Factual | Jan 2001 | The projections being discussed assume a fixed rate of growth of 3.5% per year for the next fifty years. Ray Kurzweil · Factual · Jan 2001 | Economics and markets |
| Resolved false | Ray Kurzweil | Prediction | 2001 | Within ten years of 2001, biotechnology advances will add more than a year of life expectancy every year. Ray Kurzweil · Prediction · 2001 | Science and research methods |
| Resolved false | Jeff Bezos | Factual | 1999 | Amazon.com's music store became the leading online music retailer in its first full quarter. Jeff Bezos · Factual · 1999 | Economics and markets |
“And Pro to come…”
Demis Hassabis, Google DeepMind CEO, on X 20 May 2026 Post by @demishassabis on X, 2026-05-20 (X)
Google DeepMind will release a Gemini 3.5 Pro model following Gemini 3.5 Flash.
Resolved false The record shows it did not happen or was not so.
Said for an organisation they lead or work for; it counts on both records.
What would count as true or false
No deadline given; judged as of the date of evaluation. True: Google or Google DeepMind released a model named Gemini 3.5 Pro after Gemini 3.5 Flash, available to outside users or developers (a public preview or experimental release through the Gemini app, AI Studio or Vertex AI counts). Mostly true: a Gemini 3.5 Pro was released, but under a slightly different name or in a different order from what was implied, or only to a broad paid tier. Mostly false: it was only announced, or shown to a small group of trusted testers, with no wider release. False: Google states it was cancelled or replaced (for example, it went straight to a different model with no 3.5 Pro), or a dated model lineup published well after the claim shows no 3.5 Pro release.
Evidence
- Gemini 3.5: frontier intelligence with action Google (blog.google) · 19 May 2026 Own record
“We’re also hard at work on 3.5 Pro. It's already being used internally, and we look forward to rolling it out next month.”
- Google AI Studio Google (ai.google.dev, Gemini API release notes) Own record
“**Gemini 3.6 Flash and Gemini 3.5 Flash-Lite generally available (GA)**: Released stable, production-ready versions of our latest 3.x Flash models:”
- Gemini 3.5 Pro: what happened to Google's delayed model AIWIZ Digital Marketing (aiwiz.uk) · 17 July 2026
“On 30 September a Google spokesperson told Reuters that Google no longer plans to release 3.5 Pro.”
- Google AI Studio Google AI for Developers (Gemini API release notes) · 19 May 2026
“Released `gemini-3.5-flash`, the generally available (GA) version of [Gemini 3.5 Flash](/docs/models/gemini-3.5-flash), our most intelligent model for sustained frontier performance on agentic and coding tasks.”
- Google AI Studio Google AI for Developers (Gemini API release notes) · 13 October 2026
“Released [Gemini 4 Argon](/docs/models/gemini-4-argon) (`gemini-4-argon`), our most capable model built for complex coding, agentic, math, and science tasks.”
- Google AI Studio Google AI for Developers (Gemini API release notes) · 18 September 2026
“For any new projects, use our latest models: 3.5 Flash-Lite or 3.8 Flash.”
- DD News DD News (republishing Reuters) · 4 November 2026
“Google no longer plans to release Gemini 3.5 Pro, which CEO Sundar Pichai originally said would be released in June, according to the company spokesperson.”
- Gemini 4 Carbon tops AI index; cyber defenders get it first R&D World · 28 October 2026
“When Google announced Argon on September 30, a spokesperson told Reuters that the company no longer plans to release 3.5 Pro and that Argon is larger than Google’s earlier Pro models.”
- Gemini 4 Carbon tops AI index; cyber defenders get it first R&D World · 28 October 2026
“At its I/O conference, Google promised Gemini 3.5 Pro in June. That model slipped repeatedly, and Bloomberg reported its coding performance fell short of internal goals.”
- Gemini 4 Enters Pre-Training as Gemini 3.5 Pro Stays in Testing WindowsForum · 28 July 2026
“The current status is clearer than the rumor cycle around it: Gemini 3.5 Pro is still in testing with partners.”
Reasoning
Google's Gemini 3.5 post said 3.5 Pro would roll out the next month (E1), and the API release notes record the release of 3.5 Flash on May 19, 2026 (E4). Independent reports say the model slipped repeatedly and was still in partner testing in July (E10, E11). Reuters, republished by DD News, and R&D World both report that a Google spokesperson said the company no longer plans to release Gemini 3.5 Pro. Google announced Gemini 4 Argon on September 30, when a spokesperson said 3.5 Pro was no longer planned, and released Argon in October (E9, E8, E6). This meets the criterion for cancellation or replacement, so the claim resolved false.
Cite this claim
Demis Hassabis, “Google DeepMind will release a Gemini 3.5 Pro model following Gemini 3.5 Flash.” (said 20 May 2026). Verdict: resolved false. Reliability API, claim c_9ac02f397a48, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_9ac02f397a48
Verdicts are made automatically; how verdicts and scores are made. The same data is on Hugging Face under CC BY 4.0.