{
 "episode_id": "all-in-2026-05-01-openai-misses-targets-codex-vs-claude-elon-vs-sam-trial-big-",
 "title": "OpenAI Misses Targets, Codex vs Claude, Elon vs Sam Trial, Big Hyperscaler Beats, Peptide Craze",
 "podcast": "all-in",
 "published_at": "2026-05-01",
 "speakers": [
  "David Friedberg",
  "Jason Calacanis",
  "Chamath Palihapitiya",
  "David Sacks"
 ],
 "summary": "The All-In hosts open on a Wall Street Journal report that OpenAI missed both its end-of-2025 billion-weekly-active-user and revenue targets, then debate whether the real constraint is demand or the power/compute supply needed to generate tokens; Sacks argues OpenAI's GPT-5.5 and Codex are pulling ahead while Anthropic's Opus 4.7 disappoints and is token-constrained, Chamath argues everything is power-constrained and money should follow the hyperscalers, and Friedberg invokes BCG's 'Rule of Three' plus MIT pruning research that could cut inference cost 10x. They cover the Musk v. Altman bench trial (including Greg Brockman's incriminating diary read aloud from discovery), the GPT-5.5 Cyber vs Mythos cyber race, and Big Tech's blowout earnings with ~$725B of 2026 CapEx guidance — comparing it to the 2000 Cisco/dark-fiber bust (which Sacks rejects), debating vibe-coding risk after an AI agent deleted a production database, and arguing AI is now synonymous with US economic growth and won't cause mass job loss. Friedberg then walks through the phase-3 data on Eli Lilly's retatrutide (a triple agonist) driving the peptide craze, and closes with a detailed account of his visit to a live Supreme Court hearing on the Monsanto/Roundup (FIFRA federal-preemption) case and worries about future court-packing.",
 "topics": [
  {
   "name": "Cold open: gay-name/straight-name banter",
   "timestamp": "0:00",
   "summary": "Jason plays a 'Miss Thing' podcast bit doing 'gay name or straight name' on Friedberg and Chamath; pure comedic banter before the show proper."
  },
  {
   "name": "OpenAI misses targets — power/compute as the real constraint",
   "timestamp": "2:50",
   "summary": "A WSJ report says OpenAI missed its billion-WAU and revenue targets; the hosts debate whether OpenAI can fund $600B in compute commitments, with Sacks crediting GPT-5.5/Codex and Google's share gains, Chamath arguing the binding constraint is power not demand, and Friedberg applying BCG's Rule of Three and MIT pruning research."
  },
  {
   "name": "The AI cyber race: Mythos vs GPT-5.5 Cyber",
   "timestamp": "20:30",
   "summary": "Discussion of Anthropic's Mythos and OpenAI's newly released GPT-5.5 Cyber, whether AI cyber tools are a doomsday weapon (Sacks says no — they discover, not create, vulnerabilities), the coming machines-vs-machines phase, and that Chinese models will reach the same capability within ~6 months."
  },
  {
   "name": "Musk v. Altman trial and Brockman's diary",
   "timestamp": "30:52",
   "summary": "Elon's suit accusing OpenAI of looting a charity, seeking $150B in damages and reverting to a nonprofit; the hosts marvel at Greg Brockman's incriminating discovery diary, note it's an advisory-jury bench trial before Judge Rogers, and speculate about a settlement while Polymarket holds at ~42-43% that Elon wins."
  },
  {
   "name": "Big Tech earnings, CapEx explosion, and the Cisco comparison",
   "timestamp": "41:00",
   "summary": "Google, Microsoft, Amazon and Meta beat earnings with ~$725B of 2026 CapEx guidance and collapsing free cash flow; Chamath says follow the dollars into asset-heavy suppliers, Sacks rejects the 2000 dark-fiber/Cisco analogy and argues the AI bull thesis was validated and AI is now synonymous with US growth."
  },
  {
   "name": "Vibe-coding risk and the job-loss debate",
   "timestamp": "52:14",
   "summary": "An AI agent deleted a production database (and backups) at Pocket OS; the hosts argue agents must be supervised, that 'eliminating all software developers' was peak inflated expectations, and that vibe coding is a risk that will get people fired — while AI won't cause mass job loss."
  },
  {
   "name": "Peptide craze: retatrutide phase-3 data",
   "timestamp": "58:43",
   "summary": "Friedberg breaks down Eli Lilly's retatrutide, a triple agonist (adding glucagon) showing large fat-favoring weight loss, big cholesterol/triglyceride/liver-fat/A1C improvements and anti-inflammatory/de-aging potential, with FDA approval expected ~mid-2027 and Lilly's tiered pricing strategy."
  },
  {
   "name": "Friedberg at the Supreme Court: Monsanto/Roundup and court-packing",
   "timestamp": "1:06:30",
   "summary": "Friedberg recounts attending live oral arguments in the Monsanto/Roundup FIFRA federal-preemption case (EPA-set label vs state failure-to-warn after Chevron's overturn), praises the Court as a uniquely functional institution, and Sacks warns of future court-packing to 13 justices."
  }
 ],
 "entities": [
  {
   "name": "OpenAI",
   "type": "company",
   "salience": "central",
   "role_in_episode": "Subject of the opening segment — missed user/revenue targets, $600B compute commitments, GPT-5.5/Codex momentum, the IPO question, and the Musk lawsuit defendant."
  },
  {
   "name": "Anthropic",
   "type": "company",
   "salience": "central",
   "role_in_episode": "OpenAI's main rival; framed as token/compute-constrained with a disappointing Opus 4.7 release and the Mythos cyber model, but still a multi-trillion-dollar company."
  },
  {
   "name": "Sam Altman",
   "type": "person",
   "salience": "discussed",
   "role_in_episode": "OpenAI CEO; debated whether he'll be 'right for the wrong reason' on compute bets, and a defendant Elon wants removed in the trial."
  },
  {
   "name": "Elon Musk",
   "type": "person",
   "salience": "discussed",
   "role_in_episode": "Plaintiff in the trial against OpenAI; also framed as well-positioned to use Grok/SpaceX excess compute to run through the market and potentially partner with Anthropic."
  },
  {
   "name": "Greg Brockman",
   "type": "person",
   "salience": "discussed",
   "role_in_episode": "OpenAI co-founder whose incriminating discovery diary (read aloud by Jason) is treated as a smoking gun in the Musk lawsuit; Elon seeks his removal."
  },
  {
   "name": "GPT-5.5",
   "type": "product",
   "salience": "central",
   "role_in_episode": "OpenAI's well-reviewed new model (built on a new base model 'Spud'), credited with taking coding share via Codex, plus a GPT-5.5 Cyber variant."
  },
  {
   "name": "Codex",
   "type": "product",
   "salience": "discussed",
   "role_in_episode": "OpenAI's coding tool said to be taking coding-token share from Anthropic's Opus as developer momentum shifts."
  },
  {
   "name": "Claude",
   "type": "product",
   "salience": "discussed",
   "role_in_episode": "Anthropic's model line; Opus 4.7 called a bust (compute-rationing, rollbacks to 4.6), and Claude/Cursor blamed for deleting a production database."
  },
  {
   "name": "Mythos",
   "type": "product",
   "salience": "discussed",
   "role_in_episode": "Anthropic's frontier cyber model — a not-yet-commercial proof of concept that triggered cyber panic; compared against OpenAI's GPT-5.5 Cyber."
  },
  {
   "name": "Google",
   "type": "company",
   "salience": "discussed",
   "role_in_episode": "Credited with taking meaningful consumer share via Gemini-in-search and leading/contending in enterprise (Vertex); blowout cloud earnings and a ripping stock."
  },
  {
   "name": "Chamath Palihapitiya",
   "type": "person",
   "salience": "central",
   "role_in_episode": "Host arguing the binding constraint is power, that money should follow hyperscaler CapEx into suppliers, and that vibe-coding will torch some companies."
  },
  {
   "name": "David Sacks",
   "type": "person",
   "salience": "central",
   "role_in_episode": "Host making the contrarian bull case for OpenAI's products, rejecting the dark-fiber analogy, and arguing AI is synonymous with US growth with no mass job loss."
  },
  {
   "name": "David Friedberg",
   "type": "person",
   "salience": "central",
   "role_in_episode": "Host advancing the Rule-of-Three market-structure thesis and MIT pruning argument, breaking down retatrutide, and recounting his Supreme Court visit."
  },
  {
   "name": "Jason Calacanis",
   "type": "person",
   "salience": "central",
   "role_in_episode": "Moderator framing each topic, reading Brockman's diary and Elon's quote from the trial coverage, and riffing on therapy/rumination."
  },
  {
   "name": "Hyperscalers",
   "type": "concept",
   "salience": "discussed",
   "role_in_episode": "Oracle, Amazon, Meta, Microsoft and Google — framed as the beneficiaries of the compute crunch and the destination Chamath says to follow the dollars into."
  },
  {
   "name": "Nvidia",
   "type": "company",
   "salience": "mentioned",
   "role_in_episode": "Referenced via 'no dark GPUs' (Brad Gerstner) and Friedberg showing off code-pushing in front of Jensen; not a focus."
  },
  {
   "name": "Amazon",
   "type": "company",
   "salience": "discussed",
   "role_in_episode": "Hyperscaler with the largest 2026 CapEx guidance (~$200B), AWS +28% growth, and free cash flow down 97%; flagged as the most extreme case of the asset-heavy shift."
  },
  {
   "name": "Eli Lilly",
   "type": "company",
   "salience": "central",
   "role_in_episode": "Maker of tirzepatide and the phase-3 triple-agonist retatrutide driving the peptide craze; cut a Trump-administration deal to lower tirzepatide pricing."
  },
  {
   "name": "Retatrutide",
   "type": "product",
   "salience": "central",
   "role_in_episode": "Lilly's triple-agonist (adding glucagon) with strong phase-3 fat-loss, metabolic and anti-inflammatory data; the centerpiece of the peptide segment."
  },
  {
   "name": "Monsanto",
   "type": "company",
   "salience": "discussed",
   "role_in_episode": "Roundup maker (now owned by Bayer) at the center of the Supreme Court FIFRA federal-preemption case Friedberg attended."
  },
  {
   "name": "Supreme Court",
   "type": "organization",
   "salience": "central",
   "role_in_episode": "Subject of the closing segment — Friedberg's account of live oral arguments and Sacks's warning about future court-packing."
  },
  {
   "name": "FIFRA",
   "type": "concept",
   "salience": "discussed",
   "role_in_episode": "The federal pesticide-labeling act underpinning Bayer's federal-preemption argument that the EPA, not states, sets the cancer-warning label."
  },
  {
   "name": "Chevron Doctrine",
   "type": "concept",
   "salience": "mentioned",
   "role_in_episode": "Its overturning is invoked by the plaintiff's lawyer to argue states should read the law directly rather than defer to the EPA."
  },
  {
   "name": "Polymarket",
   "type": "company",
   "salience": "mentioned",
   "role_in_episode": "Prediction market cited for OpenAI-IPO odds (~32%) and the Elon-wins odds (~42-43%) that the hosts note hasn't moved despite discovery."
  },
  {
   "name": "Cisco",
   "type": "company",
   "salience": "mentioned",
   "role_in_episode": "Invoked as the 2000-bubble cautionary chart (25 years to regain its peak) that Sacks argues does NOT apply because there are no dark GPUs."
  }
 ],
 "claims": [
  {
   "holder": "David Sacks",
   "claim": "Sam Altman will end up being right about his huge compute commitments but for the wrong reason: he missed on consumer, yet because coding became the all-important sector and OpenAI now has more compute than a token-constrained Anthropic, enterprise/Codex lets him catch up.",
   "stance": "opinion",
   "conviction": 0.7,
   "subjects": [
    "Sam Altman",
    "OpenAI",
    "Anthropic",
    "Codex"
   ],
   "timestamp": "4:49",
   "quote": "I think Sam may end up being right here for the wrong reason, which is he missed on consumer, but enterprise is going gangbusters and is giving him the ability now, I think, to catch up on Codex."
  },
  {
   "holder": "David Sacks",
   "claim": "At a product level OpenAI just had a better couple of weeks than Anthropic: GPT-5.5 reviewed strongly while Opus 4.7 appears to be a bust with people rolling back to 4.6.",
   "stance": "assertion",
   "conviction": 0.7,
   "subjects": [
    "OpenAI",
    "Anthropic",
    "GPT-5.5",
    "Claude"
   ],
   "timestamp": "4:49",
   "quote": "Opus 4.7, which is the latest Anthropic release, appears to be a bust. People are complaining about it."
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "The binding constraint across the entire AI market is the supply of power (not demand); missed forecasts reflect insufficient compute capacity, not weak demand.",
   "stance": "assertion",
   "conviction": 0.9,
   "subjects": [
    "OpenAI",
    "Anthropic",
    "Hyperscalers"
   ],
   "timestamp": "8:29",
   "quote": "Everything in this market is power constrained. The reason that these folks may miss a number or a forecast have nothing to do with demand. It is entirely 100 percent due to the supply of the power necessary to generate the output token."
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "Power/component supply-chain backlogs will hurt OpenAI and Anthropic the most and benefit the hyperscalers (Oracle, Amazon, Meta, Microsoft, Google), forcing model labs to trade equity and control for compute.",
   "stance": "prediction",
   "conviction": 0.8,
   "subjects": [
    "OpenAI",
    "Anthropic",
    "Hyperscalers"
   ],
   "timestamp": "8:29",
   "quote": "Who will this hurt? It will hurt Anthropic and OpenAI the most. Who will this benefit? It will benefit the hyperscalers, specifically Oracle, Amazon, Meta, Microsoft, and Google."
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "Elon and Dario should do a deal immediately — if the models catch up in quality, SpaceX/Grok's excess compute capacity is a huge lane Elon can run through.",
   "stance": "opinion",
   "conviction": 0.65,
   "subjects": [
    "Elon Musk",
    "Anthropic"
   ],
   "timestamp": "11:38",
   "quote": "But man, he and Dario should do a deal tomorrow."
  },
  {
   "holder": "David Friedberg",
   "claim": "The AI market will evolve toward BCG's 'Rule of Three' 4-2-1 market-share structure, splitting into a consumer market (ChatGPT/Google fighting for first, Anthropic third) and an enterprise market (Anthropic or Google leading).",
   "stance": "prediction",
   "conviction": 0.6,
   "subjects": [
    "OpenAI",
    "Google",
    "Anthropic"
   ],
   "timestamp": "13:02",
   "quote": "the consumer market looks like it's trending towards a ChatGPT slash Google fight for first place and second place, and then probably anthropic in third place"
  },
  {
   "holder": "David Friedberg",
   "claim": "Algorithmic pruning of large models (per a new MIT paper) can cut inference cost ~10x with no loss of accuracy, so AI is still in the very early days of efficiency gains.",
   "stance": "opinion",
   "conviction": 0.65,
   "subjects": [
    "OpenAI",
    "Anthropic"
   ],
   "timestamp": "13:02",
   "quote": "you can actually reduce inference costs by 10x. You can get 10x the output per energy unit that goes into the data center with no loss of accuracy"
  },
  {
   "holder": "David Sacks",
   "claim": "If there's a single reason OpenAI missed its consumer user and revenue targets, it's that Google took meaningful share after Sergey refocused the company and put Gemini at the top of search.",
   "stance": "opinion",
   "conviction": 0.75,
   "subjects": [
    "OpenAI",
    "Google"
   ],
   "timestamp": "18:37",
   "quote": "if there is a single reason why OpenAI did not hit its user targets and its revenue targets, certainly around consumer, you'd have to say it's because Google managed to take meaningful share"
  },
  {
   "holder": "David Sacks",
   "claim": "AI cyber models like Mythos and GPT-5.5 Cyber are not a doomsday weapon — they discover existing vulnerabilities rather than create them, so used by defenders first they harden infrastructure.",
   "stance": "opinion",
   "conviction": 0.8,
   "subjects": [
    "Mythos",
    "GPT-5.5"
   ],
   "timestamp": "21:48",
   "quote": "it doesn't create the vulnerabilities. It just discovers them. The bugs were already in the code."
  },
  {
   "holder": "David Sacks",
   "claim": "The shift from pre-AI to post-AI cyber will be a one-time upgrade cycle, after which a new equilibrium between AI-powered offense and defense makes it feel normal rather than existential.",
   "stance": "prediction",
   "conviction": 0.7,
   "subjects": [
    "Mythos",
    "GPT-5.5"
   ],
   "timestamp": "21:48",
   "quote": "once we get past that upgrade cycle, you're going to reach a new equilibrium between AI-powered cyber offense and AI-powered cyber defense"
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "The nature of cyber will change from humans-exploiting-humans to machines-versus-machines, and eventually all the operational software running the world will be rewritten by machines and become more secure even as the threat increases.",
   "stance": "prediction",
   "conviction": 0.75,
   "subjects": [
    "Mythos"
   ],
   "timestamp": "25:04",
   "quote": "In the next phase, it will be machines versus machines."
  },
  {
   "holder": "Greg Brockman",
   "reported_by": "Jason Calacanis",
   "claim": "Per his discovery diary, OpenAI's true aim was to push Elon out and convert to a for-profit, and the public story would correctly be that they weren't honest with him.",
   "stance": "assertion",
   "conviction": 0.9,
   "subjects": [
    "OpenAI",
    "Elon Musk",
    "Greg Brockman"
   ],
   "timestamp": "30:52",
   "quote": "The true answer is that we want Elon out. If three months later we're doing B Corp, then it was a lie."
  },
  {
   "holder": "Elon Musk",
   "reported_by": "Jason Calacanis",
   "claim": "Allowing OpenAI to convert from a charity to a for-profit would destroy the entire foundation of charitable giving in America.",
   "stance": "assertion",
   "conviction": 0.9,
   "subjects": [
    "OpenAI",
    "Elon Musk"
   ],
   "timestamp": "30:52",
   "quote": "if we make it okay to loot a charity, the entire foundation of charitable giving in America will be destroyed"
  },
  {
   "holder": "Jason Calacanis",
   "claim": "The most likely outcome of the Musk v. Altman trial is some kind of settlement rather than a full unwinding.",
   "stance": "prediction",
   "conviction": 0.5,
   "subjects": [
    "OpenAI",
    "Elon Musk",
    "Sam Altman"
   ],
   "timestamp": "37:45",
   "quote": "my guess is, like most of these things, there'll be some sort of settlement or something here, but maybe he takes it to the mat"
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "Both OpenAI and Anthropic are multi-trillion-dollar companies and will be fine despite the missed targets.",
   "stance": "opinion",
   "conviction": 0.8,
   "subjects": [
    "OpenAI",
    "Anthropic"
   ],
   "timestamp": "8:29",
   "quote": "I think this is a multi-trillion dollar company. I think Anthropic is a multi-trillion dollar company."
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "The capital markets are undergoing a structural shift from asset-light to asset-heavy; hyperscalers will get levered and look like bulky industrial businesses in five years with no good valuation case, so investors should follow the trillion dollars of CapEx into the underpriced suppliers.",
   "stance": "prediction",
   "conviction": 0.75,
   "subjects": [
    "Hyperscalers"
   ],
   "timestamp": "43:30",
   "quote": "they're going to look like this big bulky industrial business in five years. I'm not sure that there's a good valuation case to be made at that point."
  },
  {
   "holder": "David Sacks",
   "claim": "This CapEx boom is not a repeat of the 2000 dark-fiber/Cisco bust because there are no idle 'dark GPUs' — voracious demand for compute and tokens is pulling the investment forward and just validated the AI bull thesis.",
   "stance": "opinion",
   "conviction": 0.85,
   "subjects": [
    "Cisco",
    "Hyperscalers",
    "Nvidia"
   ],
   "timestamp": "46:55",
   "quote": "There's no dark GPUs today, as Brad Gerstner likes to say."
  },
  {
   "holder": "David Sacks",
   "claim": "AI will not cause the huge job loss the doomers predict; it's becoming synonymous with the American economy and with economic growth, so halting AI means halting the economy.",
   "stance": "opinion",
   "conviction": 0.85,
   "subjects": [
    "Hyperscalers"
   ],
   "timestamp": "49:38",
   "quote": "I don't think there's going to be this huge job loss associated with this productivity boom that we're going to get."
  },
  {
   "holder": "David Sacks",
   "claim": "AI agents are valuable but must be supervised and accountable to people; the idea of automating all jobs away hand-waves over real technical problems, and 'eliminating all software developers' was the peak of inflated expectations.",
   "stance": "opinion",
   "conviction": 0.8,
   "subjects": [
    "Claude"
   ],
   "timestamp": "47:35",
   "quote": "this idea of eliminating all software developers was the peak of inflated expectations"
  },
  {
   "holder": "David Sacks",
   "claim": "The Pocket OS database deletion was old-fashioned bugs at an edge case, not AI scheming; the systemic issue is that AI still doesn't know what it doesn't know and can be overconfident, which is why it must be supervised.",
   "stance": "assertion",
   "conviction": 0.75,
   "subjects": [
    "Claude"
   ],
   "timestamp": "54:57",
   "quote": "AI still doesn't know what it doesn't know. You know, like a human would stop before deleting a production database"
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "Vibe coding is a serious risk, not just a tax — some public-company employees will torch enterprise value trying to vibe-code their way out of something and will get fired.",
   "stance": "prediction",
   "conviction": 0.8,
   "subjects": [
    "Claude"
   ],
   "timestamp": "57:45",
   "quote": "People will get fired because there will be some public companies where some goofball tries to vibe code their way out of something and they're going to torch the enterprise value."
  },
  {
   "holder": "David Friedberg",
   "claim": "Retatrutide's added glucagon agonist favors fat burning over muscle burning, which could make it useful even at low dose for cutting weight while maintaining muscle, and it may double as an anti-inflammatory/de-aging drug beyond obesity and diabetes.",
   "stance": "opinion",
   "conviction": 0.7,
   "subjects": [
    "Retatrutide",
    "Eli Lilly"
   ],
   "timestamp": "1:00:59",
   "quote": "reticulotide might actually be kind of a de-aging drug as well"
  },
  {
   "holder": "David Friedberg",
   "claim": "Lilly will use tiered pricing — cheap tirzepatide (~$50 on Medicare) as the 'Honda' and premium retatrutide as the 'Mercedes' upgrade — to make most of its money on the high-end product.",
   "stance": "opinion",
   "conviction": 0.7,
   "subjects": [
    "Eli Lilly",
    "Retatrutide"
   ],
   "timestamp": "1:03:46",
   "quote": "you get terzepotide for 50 bucks. But if you want to upgrade, get the Retatrutide, that's the high premium product. And that's where they're going to start to make all this money. That'll be the Mercedes to the Honda."
  },
  {
   "holder": "David Friedberg",
   "claim": "If the Supreme Court rules states can interpret federal law and ignore federal regulatory bodies, it opens a can of worms letting states ignore the EPA, FDA, USDA and others.",
   "stance": "opinion",
   "conviction": 0.7,
   "subjects": [
    "Supreme Court",
    "FIFRA",
    "Monsanto"
   ],
   "timestamp": "1:13:22",
   "quote": "if the states get to interpret federal law and ignore federal regulatory bodies, it opens up a whole new can of worms in terms of like all the states can start to ignore federal regulatory bodies, like the EPA or the FDA or the USDA"
  },
  {
   "holder": "David Sacks",
   "claim": "The Supreme Court is one of the last highly functional institutions in the US, but court-packing will eventually expand it to some large number of justices, so enjoy it in its current form while it lasts.",
   "stance": "prediction",
   "conviction": 0.7,
   "subjects": [
    "Supreme Court"
   ],
   "timestamp": "1:17:07",
   "quote": "I think the Supreme Court is one of the last highly functional institutions in the United States."
  },
  {
   "holder": "David Sacks",
   "claim": "Public approval of the Supreme Court is driven by whether people agree with its recent decisions, not by the quality of its process or argumentation.",
   "stance": "opinion",
   "conviction": 0.75,
   "subjects": [
    "Supreme Court"
   ],
   "timestamp": "1:17:45",
   "quote": "The popularity of the court really depends on whether it's issuing decisions that people agree with. That's what it comes down to."
  },
  {
   "holder": "Chamath Palihapitiya",
   "claim": "Roughly 40% of all announced AI/data-center projects will get canceled because the industry did such a poor job creating a positive halo around AI, mirroring that 40% of all projects in the last four years got canceled.",
   "stance": "prediction",
   "conviction": 0.6,
   "subjects": [
    "Hyperscalers",
    "OpenAI",
    "Anthropic"
   ],
   "timestamp": "12:03",
   "quote": "40% of all the announced projects get canceled because 40% of all projects in the last four years have been canceled"
  }
 ],
 "predictions": [
  {
   "holder": "David Sacks",
   "prediction": "Within about six months all the frontier models will have Mythos-level cyber capability.",
   "timeframe": "within ~6 months",
   "subjects": [
    "Mythos",
    "GPT-5.5"
   ],
   "timestamp": "21:20",
   "quote": "within six months or so, all the frontier models are going to have Mythos level cyber capability",
   "due": "2026-11-01"
  },
  {
   "holder": "David Sacks",
   "prediction": "Chinese models will reach frontier cyber capability within approximately six months.",
   "timeframe": "within ~6 months",
   "subjects": [
    "Mythos"
   ],
   "timestamp": "24:05",
   "quote": "the Chinese models are going to have these capabilities within approximately six months",
   "due": "2026-11-01"
  },
  {
   "holder": "Chamath Palihapitiya",
   "prediction": "Probably within the next five or six years, all the operational software that runs the world will get rewritten, increasingly by machines.",
   "timeframe": "next 5-6 years",
   "subjects": [
    "Mythos"
   ],
   "timestamp": "25:04",
   "quote": "Probably in the next five or six years, there'll be so much reason to rewrite all of the software that runs the world.",
   "due": "2032-05-01"
  },
  {
   "holder": "Chamath Palihapitiya",
   "prediction": "The hyperscalers will spend ~$700B/year of CapEx, get heavily levered, and look like bulky industrial businesses with no good valuation case in five years.",
   "timeframe": "within 5 years",
   "subjects": [
    "Hyperscalers"
   ],
   "timestamp": "43:30",
   "quote": "they're going to look like this big bulky industrial business in five years",
   "due": "2031-05-01"
  },
  {
   "holder": "David Friedberg",
   "prediction": "The AI market will settle into a 4-2-1 'Rule of Three' market-share structure across separate consumer and enterprise markets.",
   "timeframe": "unspecified",
   "subjects": [
    "OpenAI",
    "Google",
    "Anthropic"
   ],
   "timestamp": "13:02",
   "quote": "any stable, mature, competitive market evolves to a 4 to 2 to 1 ratio of market share for basically 90% of the market"
  },
  {
   "holder": "David Friedberg",
   "prediction": "Retatrutide could be approved by the FDA around mid-2027 (possibly sooner).",
   "timeframe": "mid-2027",
   "subjects": [
    "Retatrutide",
    "Eli Lilly"
   ],
   "timestamp": "1:01:24",
   "quote": "could happen sooner, could happen sometime later this year",
   "due": "2027-06-30"
  },
  {
   "holder": "Chamath Palihapitiya",
   "prediction": "About 40% of all announced AI/data-center projects will get canceled.",
   "timeframe": "unspecified",
   "subjects": [
    "Hyperscalers",
    "OpenAI",
    "Anthropic"
   ],
   "timestamp": "12:03",
   "quote": "40% of all the announced projects get canceled"
  },
  {
   "holder": "David Sacks",
   "prediction": "There will be over $700B of hyperscaler CapEx this year (up from ~$350B last year), more than 2% of GDP.",
   "timeframe": "2026",
   "subjects": [
    "Hyperscalers"
   ],
   "timestamp": "46:55",
   "quote": "This year, I think we were supposed to have 660 billion of hyperscaler CapEx up from 350 last year. I think there's now the new estimate is going to be over 700",
   "due": "2026-12-31"
  },
  {
   "holder": "David Sacks",
   "prediction": "Court-packing will eventually expand the Supreme Court to 13 (or more) justices.",
   "timeframe": "unspecified",
   "subjects": [
    "Supreme Court"
   ],
   "timestamp": "1:17:07",
   "quote": "at some point, we're going to have like 13 or 21 or some crazy number of justices up there"
  }
 ],
 "facts": [
  {
   "statement": "A WSJ investigative report said OpenAI expected to hit 1 billion weekly active users before the end of 2025 but missed it and still hadn't hit it four months into 2026, and also missed its 2025 ChatGPT revenue target.",
   "subjects": [
    "OpenAI"
   ],
   "value": "1B WAU target missed",
   "timestamp": "2:50",
   "quote": "OpenAI expected to hit 1 billion WoWs weekly active users before the end of 2025 They missed that, and they still haven't hit the milestone four months into 2026",
   "source": "Jason Calacanis"
  },
  {
   "statement": "OpenAI's ChatGPT is at roughly a $20-30 billion run rate.",
   "subjects": [
    "OpenAI"
   ],
   "value": "$20-30B run rate",
   "timestamp": "2:50",
   "quote": "they're at a 20, 30 billion dollar run rate",
   "source": "Jason Calacanis"
  },
  {
   "statement": "OpenAI has about $600 billion in compute spending commitments, roughly equal to its secondary-market valuation.",
   "subjects": [
    "OpenAI"
   ],
   "value": "$600B compute commitments",
   "timestamp": "2:50",
   "quote": "OpenAI has $600 billion in spending commitments for Compute",
   "source": "Jason Calacanis"
  },
  {
   "statement": "OpenAI still has about 900 million weekly users, well ahead of Claude (estimated sub-100 million), with Gemini in the 700M-1B range.",
   "subjects": [
    "OpenAI",
    "Claude",
    "Google"
   ],
   "value": "~900M weekly users",
   "timestamp": "13:02",
   "quote": "they're still at 900 million weekly users, which is well ahead of whatever Claude is at. I think Claude is probably sub 100 million",
   "source": "David Friedberg"
  },
  {
   "statement": "Polymarket showed a 32% chance OpenAI goes public by the end of 2026, down from 60% in December.",
   "subjects": [
    "OpenAI",
    "Polymarket"
   ],
   "value": "32% (down from 60%)",
   "timestamp": "7:46",
   "quote": "Polymarket showing now a 32% chance that OpenAI goes public by the end of 2026 This is down from 60% in December.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "Google claims 75% of GCP customers are active users of Vertex.",
   "subjects": [
    "Google"
   ],
   "value": "75% of GCP customers",
   "timestamp": "13:02",
   "quote": "Google claims that 75% of GCP customers are active users of Vertex",
   "source": "David Friedberg"
  },
  {
   "statement": "GPT-5.5 Cyber is the second model (tested by the AI Security Institute) to complete a multi-step cyber attack simulation end-to-end, with Mythos-level capability but commercially ready.",
   "subjects": [
    "GPT-5.5",
    "Mythos"
   ],
   "value": "2nd model to finish end-to-end",
   "timestamp": "18:37",
   "quote": "GPT-55 is the second model to complete one of their multi-step cyber attack simulations end to end. It has the same level of capability as Mythos",
   "source": "David Sacks"
  },
  {
   "statement": "Elon Musk's suit accuses OpenAI of breach of charitable trust and unjust enrichment, seeks $150 billion in damages, asks that OpenAI revert to a nonprofit, and that Altman and Brockman be removed.",
   "subjects": [
    "OpenAI",
    "Elon Musk",
    "Sam Altman",
    "Greg Brockman"
   ],
   "value": "$150B in damages sought",
   "timestamp": "30:52",
   "quote": "He's seeking $150 billion in damages that they revert back to a non-profit that Altman and Brockman be removed.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "Polymarket has stayed at roughly 42-43% that Elon wins, unmoved even after the discovery documents were published.",
   "subjects": [
    "Polymarket",
    "Elon Musk",
    "OpenAI"
   ],
   "value": "42-43%",
   "timestamp": "34:28",
   "quote": "It's effectively at 42 or 43 percent that Elon wins.",
   "source": "Chamath Palihapitiya"
  },
  {
   "statement": "The Musk v. Altman case is a bench trial before Judge Rogers (a 61-year-old Obama appointee) where the jury is advisory and the judge makes the final call and sets damages.",
   "subjects": [
    "OpenAI",
    "Elon Musk"
   ],
   "value": "bench trial, advisory jury",
   "timestamp": "36:13",
   "quote": "It's a bench trial where the jury is advisory in capacity. But ultimately, that judge, she will make the final call.",
   "source": "Chamath Palihapitiya"
  },
  {
   "statement": "Google, Microsoft, Amazon and Meta guided to ~$725 billion of 2026 CapEx — Amazon ~$200B, Microsoft and Google ~$190B each, Meta ~$145B.",
   "subjects": [
    "Hyperscalers",
    "Google",
    "Amazon"
   ],
   "value": "$725B 2026 CapEx",
   "timestamp": "41:00",
   "quote": "Amazon leading the pack with $200 billion, $190 billion each for Microsoft and Google, $145 billion for Meta.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "Cloud growth was strong: Google Cloud +63% on $20B, Microsoft Cloud +30% on $34.7B, AWS +28% on $37.6B (quarterly).",
   "subjects": [
    "Google",
    "Hyperscalers",
    "Amazon"
   ],
   "value": "GCP +63%, MS +30%, AWS +28%",
   "timestamp": "41:00",
   "quote": "Google Cloud, which includes the Google Suite, that grew 63% year on year. Let that number sink in. 63% on $20 billion in revenue.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "Hyperscaler free cash flow collapsed: Amazon down 97%, with Google, Microsoft and Meta down 12%, 12% and 8% respectively.",
   "subjects": [
    "Hyperscalers",
    "Amazon"
   ],
   "value": "Amazon FCF -97%",
   "timestamp": "41:00",
   "quote": "Amazon's free cashflow down 97 percent, Google, Microsoft, and Meta down 12, 12 and 8 percent respectively.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "Microsoft's power purchase agreement to restart Three Mile Island was for more than 2x the prevailing spot rate for energy.",
   "subjects": [
    "Hyperscalers"
   ],
   "value": ">2x spot rate",
   "timestamp": "43:30",
   "quote": "when Microsoft convinced the owners of Three Mile Island to turn their nuclear site back on, do you know what their Ford Purchase Agreement was? It was for more than 2x to prevailing spot rate for energy.",
   "source": "Chamath Palihapitiya"
  },
  {
   "statement": "It took Cisco about 25 years to get back to its 2000 peak.",
   "subjects": [
    "Cisco"
   ],
   "value": "~25 years to recover",
   "timestamp": "46:03",
   "quote": "It took them 25 years to get back to that peak.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "AI was about 75% of GDP growth in the last quarter, and CapEx alone is more than 2% of GDP.",
   "subjects": [
    "Hyperscalers"
   ],
   "value": "75% of GDP growth",
   "timestamp": "46:55",
   "quote": "I think in the last quarter, AI was 75 percent of GDP growth",
   "source": "David Sacks"
  },
  {
   "statement": "An AI coding agent (Opus 4.6 via Cursor) deleted Pocket OS's production database, including backups, by deleting a railway volume without user confirmation.",
   "subjects": [
    "Claude"
   ],
   "value": "production DB + backups deleted",
   "timestamp": "52:14",
   "quote": "they decided to fix the mismatch by deleting a railway volume without user confirmation. And they pushed the code from a repo to a live app, and they deleted everything, including the backups.",
   "source": "Jason Calacanis"
  },
  {
   "statement": "In retatrutide's phase-3 trial the average user's weight fell from 214 lbs, losing 37 lbs versus 6 lbs on placebo over 40 weeks.",
   "subjects": [
    "Retatrutide",
    "Eli Lilly"
   ],
   "value": "-37 lbs vs -6 lbs placebo",
   "timestamp": "59:57",
   "quote": "The average user in this phase three trial saw their weight decline from 214 lbs. They lost 37 lbs. That's compared to 6 lbs. on placebo in 40 weeks.",
   "source": "David Friedberg"
  },
  {
   "statement": "Retatrutide data showed non-HDL cholesterol down 27%, triglycerides down 41%, liver fat down 80%, and A1C dropping from 7.9% to 6% in 40 weeks.",
   "subjects": [
    "Retatrutide"
   ],
   "value": "liver fat -80%, A1C 7.9%→6%",
   "timestamp": "59:57",
   "quote": "non-HDL cholesterol down 27%, triglycerides down 41%",
   "source": "David Friedberg"
  },
  {
   "statement": "In November 2025 Lilly cut a deal with the Trump administration to drop tirzepatide's price to about $50 on Medicare.",
   "subjects": [
    "Eli Lilly"
   ],
   "value": "~$50 on Medicare",
   "timestamp": "1:03:36",
   "quote": "On November of 2025, Lilly cut a deal with the Trump administration. I saw this to drop the price on terzepotide pretty significantly. I think it's like 50 bucks on Medicare.",
   "source": "David Friedberg"
  },
  {
   "statement": "Bayer (which now owns Monsanto) has paid out $10 billion in Roundup lawsuits, reserved another $10 billion, and still has 90,000 cases outstanding.",
   "subjects": [
    "Monsanto"
   ],
   "value": "$10B paid, 90,000 cases",
   "timestamp": "1:10:14",
   "quote": "Bayer, which now owns Monsanto, has paid out $10 billion in these lawsuits. And they have reserved $10 billion on their balance sheet. They have 90,000 cases still outstanding in the courts.",
   "source": "David Friedberg"
  },
  {
   "statement": "Supreme Court favorability is relatively high — about 44-50%, up from a low of 35% in the 2024 Gallup survey.",
   "subjects": [
    "Supreme Court"
   ],
   "value": "44-50% favorability",
   "timestamp": "1:18:00",
   "quote": "I think that it got as low as 35% in the 2024 Gallup Survey, so I think it's back up to 44% to 50% favorability",
   "source": "David Sacks"
  }
 ],
 "notable_quotes": [
  {
   "speaker": "Chamath Palihapitiya",
   "timestamp": "11:50",
   "quote": "Everything in this market is power constrained. The reason that these folks may miss a number or a forecast have nothing to do with demand. It is entirely 100 percent due to the supply of the power necessary to generate the output token.",
   "why": "The episode's central macro thesis — that power, not demand or even compute, is the true bottleneck on AI."
  },
  {
   "speaker": "David Sacks",
   "timestamp": "46:55",
   "quote": "There's no dark GPUs today, as Brad Gerstner likes to say.",
   "why": "The crux of the bull case rejecting the 2000 dark-fiber/Cisco-bust analogy for today's CapEx boom."
  },
  {
   "speaker": "David Sacks",
   "timestamp": "47:35",
   "quote": "the bull thesis for AI just got validated in a single afternoon",
   "why": "Memorable framing of how the hyperscalers' simultaneous earnings beats reset the AI narrative."
  },
  {
   "speaker": "David Friedberg",
   "timestamp": "32:27",
   "quote": "I just don't know why Greg Brockman's got a freaking diary where he's literally documenting.",
   "why": "Captures the table's incredulity at the smoking-gun discovery diary in the Musk v. Altman trial."
  },
  {
   "speaker": "David Sacks",
   "timestamp": "32:45",
   "quote": "It's not just journal maxing, it's discovery maxing. It's a smoking gun maxing.",
   "why": "The most-quotable line of the legal segment, coining how Brockman's diary became evidence."
  },
  {
   "speaker": "Jason Calacanis",
   "timestamp": "33:57",
   "quote": "Start new projects, 9 out of 10 foul, place 9 out of 10 bets, one wins and you're golden.",
   "why": "Jason's compressed life/investing philosophy delivered as anti-rumination advice."
  }
 ],
 "extracted_by": "agent",
 "prompt_version": "episode-extract@3",
 "_file": "all-in-2026-05-01-openai-misses-targets-codex-vs-claude-elon-vs-sam-trial-big-.json"
}