Aug 9, 2025 · 2h 26m · tbpn
Weekly Recap: Apple Sitting on $100B, Disney Enters The AI Race, Marc Andreessen, All About GPT-5
John Coogan · 1h 19m spoken Jordi Hays · 25m spoken Marc Andreessen · 20m spoken Rahul Sonwalkar · 2m spoken
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
John Coogan and Jordan Hays analyze Big Tech cloud earnings, Apple's capital allocation strategy, entertainment disruption, and the GPT-5 launch alongside guest interviews with Marc Andreessen and Rahul Sonwalkar. The episode explores the unit economics of enterprise cloud infrastructure, geopolitical open-source AI dynamics, and the commercialization of foundation models.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The hosts hold 60.4% of the talking time here. How this is scored →
The hosts as informed peer 5.6 Guest teaching 2.9 Guest disagreement 1.6 The hosts pushing back 2.0
speaking balance: gold is the hosts, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 1:19:15 Andreessen directly rejects host premise on early iPhone quality
Andreessen bluntly refutes Coogan's claim that the original iPhone was primarily an effective phone, explaining that for its first two years it suffered severe call reliability issues and antenna flaws.
Hardest push from the hosts ▶ 3:25 Host immediately halts co-host to correct cloud category errorCoogan immediately interrupts Hays's assertion about SaaS revenue to clarify the precise technical distinction between Software-as-a-Service and Infrastructure-as-a-Service in cloud earnings.
Biggest teaching moment ▶ 1:24:25 Andreessen educates on open weights vs open training corpusAndreessen explains the hidden architectural risk of open-weights models lacking open training data, highlighting international data sovereignty and unobservable alignment constraints.
The host holds their own ▶ 2:03:50 Host dismantles consumer outrage narrative using platform historyCoogan counters the narrative of Reddit outrage over OpenAI's automated routing by citing historic historical precedents from Facebook's news feed launch and early YouTube redesigns.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The hosts as informed peer | Guest teaching | Guest disagreement | The hosts pushing back | Why |
|---|---|---|---|---|---|---|
| Microsoft Earnings and Cloud Infrastructure vs. AI Growth | 7 | 2 | 2 | 4 | Coogan corrects Hays on cloud taxonomy, differentiating SaaS, PaaS, and IaaS when analyzing Azure's earnings breakdown. He demonstrates deep technical understanding of cloud margins and AI compute infrastructure workloads. | |
| Figma IPO Celebration and Partnership Shoutout | 6 | 2 | 1 | 2 | Coogan articulates game theory and Nash equilibrium to explain the duopoly pricing behavior of cloud hyperscalers and soft drink giants. The co-hosts banter playfully about Warren Buffett's beverage habits. | |
| Enterprise Cloud Rebound and Microsoft's $4 Trillion Valuation | 5 | 1 | 0 | 1 | Coogan reviews Microsoft's premium valuation multiple relative to Amazon and Google. The dialogue is collaborative financial analysis with both hosts agreeing on Azure's fundamentals. | |
| Disney's AI Dilemma: The Moana Digital Double Controversy | 6 | 1 | 0 | 1 | Coogan provides in-depth analysis on Hollywood production logistics, deepfake tech, and IP copyright liability regarding Disney's Moana production. Hays agrees and jokes along. | |
| Generative AI in Entertainment: Content Customization and Studio Economics | 6 | 3 | 3 | 4 | Hays argues that AI content customization will revolutionize storytelling, while Coogan pushes back, arguing that great entertainment relies on auteur vision rather than personalized novelty. | |
| Brainstorming Wild Uses for Apple's $100 Billion Cash Pile | 6 | 1 | 1 | 2 | Coogan riffs through creative scenarios for Apple deploying its $100B cash reserve, displaying domain knowledge on supply chains, patents, and retail footprint economics. | |
| Ben Thompson's Apple AI Analysis and Tim Cook's Operational Legacy | 6 | 1 | 0 | 1 | Coogan defends Tim Cook's operational track record and supply chain focus against critics pushing Apple to compete head-on in foundation model training. | |
| SiriusXM Cancels Howard Stern: Talent Economics in Modern Audio | 5 | 2 | 1 | 1 | Coogan and Hays analyze SiriusXM's business model, valuation multiple, and talent payout structure following Howard Stern's contract cancellation. | |
| OpenAI Releases GPT OSS: Open Weights Reasoning Models | 6 | 2 | 1 | 2 | Coogan and Tyler discuss Sam Altman's announcement of GPT OSS, analyzing open-weights reasoning architectures and debating potential distillation strategies. | |
| US-China Open Source Race: Training Cost Efficiency and Bounties | 5 | 3 | 2 | 2 | Tyler explains the training cost efficiencies of OpenAI's open models compared to DeepSeek, schooling Coogan on GPU hour math from published model cards. | |
| Marc Andreessen on AI Breakthroughs, Productization, and Lab Structures | 5 | 6 | 3 | 2 | Andreessen playfully corrects Coogan's Facebook motto phrasing to 'move fast and make things' and explains organizational friction between research scientists and product teams across major AI labs. | |
| Marc Andreessen on Hardware Obsolescence, Wearables, and Apple's Playbook | 5 | 6 | 4 | 3 | When Coogan asserts the original iPhone was primarily a phone, Andreessen firmly pushes back by recounting how early iPhones could not reliably make calls and explains how products become obsolete at the point of perfection. | |
| Open Weights vs. Open Data: Marc Andreessen on Geopolitics and Trust | 5 | 5 | 2 | 2 | Andreessen provides a nuanced breakdown of the difference between open source, open weights, and open corpus, explaining global cultural and legal concerns around embedded biases. | |
| Marc Andreessen on Ad-Supported AI, Privacy Rights, and Legal Battles | 4 | 5 | 2 | 1 | Andreessen details how ad-supported business models ensure broad global access and outlines how historical Constitutional Fourth Amendment precedents will apply to AI conversational data. | |
| Lightning Round with Marc Andreessen: AI Humor, Breaking into VC, and Antitrust | 4 | 5 | 3 | 1 | Andreessen explains the survivorship bias behind antitrust regulators taking victory laps over Figma's IPO, comparing central planning fallacies to European bottle cap regulations. | |
| GPT-5 Launch Roster: OpenAI Team Breakdown and the Competitive Landscape | 7 | 1 | 0 | 1 | Coogan delivers a rapid-fire sports-style roster breakdown of OpenAI leadership and competing labs, demonstrating detailed knowledge of key industry personnel and benchmarks. | |
| ARC-AGI Benchmark Scores and the Debut of TBPN Bench | 6 | 3 | 2 | 3 | Tyler breaks down ARC-AGI benchmark results, and Coogan debates the economic value of task-specific benchmark optimization versus general intelligence capabilities. | |
| Julius AI Founder Rahul Sonwalkar on Application Layer Economics | 5 | 4 | 1 | 2 | Sonwalkar explains Julius AI's pricing strategy and unit economics following inference cost drops. Coogan engages with practical enterprise workflow comparisons. | |
| GPT-5 User Reactions: The Model Picker Removal and Reddit Backlash | 6 | 3 | 3 | 4 | Hays highlights community backlash over OpenAI removing the model picker, while Coogan vigorously pushes back, comparing Reddit complaints to early resistance against algorithmic feeds. | |
| The Future of Scaled AI: Continuous Deployment and Capability Underhang | 7 | 2 | 1 | 2 | Coogan lays out a comprehensive thesis contrasting Apple's event-driven upgrade marketing with continuous cloud AI deployment, arguing that application-layer harnessing represents the next major S-curve. |
Statements from this episode (37)
Hays: GitHub Copilot reached around $500M in ARR
“Last we heard it was at Around half a billion in ARR.”
Coogan: Non-AI core infrastructure drove Microsoft Azure revenue growth
“More than half of Azure's 33% revenue jump in the company's March quarter came from non-AI services. While the company didn't give a comparable breakdown of the cloud unit's 39% growth in its June quarter, it said that the core infrastructure business, Microso…”
Coogan: Anthropic's next major cluster will run on Amazon
“I believe Anthropic is pretty tightly hitched to Amazon, and I think the next big cluster from Anthropic will be powered by Amazon for the most part”
Hays: Microsoft did more layoffs in 2025 than 2022-2024 combined
“One thing that stood out to me that has stood out to me across this year with Microsoft is they've done more layoffs this year than the past three years before that combined. So, twenty-twenty-two, 23, and twenty-twenty-four.”
Coogan: Amazon lags in AI partnerships and frontier model labs
“Amazon's just not there either on the partnership side or on the core like training frontier lab side.”
Coogan: OpenAI generates around $1B per month in revenue
“Like, you know, they're generating, what, a billion dollars a month in revenue at this point. Everyone uses the app. It's installed everywhere. Like, that, that, that token demand is not going anywhere, that infrastructure demand is not going anywhere, so.”
Coogan: Azure Non-AI margins are 73% vs 30-40% for AI
“Non-AI gross margins within Azure were around 73%. Wow. That compares to 30 to 40% gross margin for AI, he estimated because of the huge cost of setting up AI infrastructure.”
Coogan: Duopolies avoid price wars via Nash equilibrium marketing competition
“It's not, it is unspoken, but it is a nature, it is a natural game theoretic Nash equilibrium. Like, it is the natural state of things that both sides understand that to go to war would be mutually assured destruction, and so they don't even need to talk about…”
Coogan: Disney will win AI copyright revenue through the courts
“And I think that they will through the courts, especially because as like big companies, like you're not, so it's not this crazy, oh, there's like a bunch of kids doing random things.”
Hays: AI content customization will take off via interactive film variations
“I think that broadly, like, content customization will probably take off, because you could, Disney can make a film now, and then you could make millions of different variations of it that become interactive with, like, the underlying fan.”
Hays: Disney's ecosystem moat makes its IP far harder to replicate
“It's hard, extremely difficult and time intensive to create valuable IP. And Disney's advantage is they have this sort of, like, three 60, and that they can make a film, and they can bring it to Disneyland, they can bring it to a cruise, they can create physic…”
Coogan: Fully AI-generated custom content won't be Disney's domain
“And so I do think that the fully AI generated, fully custom content that will exist, but it will exist on independent third-party platforms. It won't be the domain of Disney.”
Coogan: Hollywood, Like Venture Capital, Requires Embracing Massive Flops
“The nature of the, of Hollywood is like venture capital. Like it is a hits driven business. If you don't accept that you're going to have massive flops and egg on your face and embarrassing moments, you will not make it.”
Coogan: Apple's core priority is hardware supply chain, not AI
“He's solving the most important problem for that company, which is not the application layer in artificial intelligence, it's the supply chain. And keeping the flows, flow of iPhones.”
Hays: SiriusXM generated $8.6B in 2024 revenue, trading below 1x
“So they did 8.6 billion in 20 24 revenue. So they're trading at less than one X revenue, which just says when you have a shrinking business, it is a rough place to be in.”
Hays: An o3-level reasoning model was open-sourced within a year of o1
“So less than one year between O one announced, which was September of 20, 24. And we have an O three level model open sourced that's runnable on consumer hardware, wild progress.”
Hays: Reflection AI is in talks to raise $1B for open-source models
“It was reported yesterday by the misinformation that reflection, reflection the misinformation reported that reflection one-year-old startup we've had the CEO on before he's in talks to raise one billion to develop open source models to compete with deep sea m…”
Andreessen: Google could have built a GPT-4-level ChatGPT by 2019
“Google developed the transformer in 2017. And then they basically let it sit on the shelf, right? Cause it was a research project. They didn't productize it. They were very worried about, you know, from people I've talked to, they were very worried about the, …”
Andreessen: Musk eliminated research vs product division at xAI
“Within XAI, he's now collapsed, you know, he's eliminated the distinction between research and product.”
Andreessen: Emulating Apple's last-mover strategy suffers from survivorship bias
“I think there's a lot of survivorship bias in these kinds of strategy discussions where people look at the one company That's able to pull this off, and they don't look at the 50 other companies that are in the graveyard, you know, because they, you know, beca…”
Andreessen: Tech products become obsolete the moment they achieve perfection
“I think technology products become obsolete at the precise moment they become perfect. And what I mean by perfect basically is like, yeah, it's like the perfect idealized complete product. Like it does everything you could possibly ever imagine, everything a c…”
Andreessen: Open-source AI lags leading proprietary models by about six months
“Right now, you know, which I think what you would say is it kind of lags the leading edge proprietary implementations by, you know, six months or something like that.”
Andreessen: DeepSeek and most open-weights AI models do not share training data
“What most of the open source open weights models don't have, including, you know, deep seek specifically, What they don't have is they don't have open data, right? Or open corpus, right? So you can't actually see the training data that went, went into them.”
Andreessen: Open-source AI developers will increasingly publish training datasets
“And then correspondingly, I think what we'll see is more open source developers also doing open corpus open data.”
Andreessen: AI cannot scale to billions of users on paid subscriptions
“If you really want to get to fi, if you want to get to a billion and then five billion people you can't do that with a paid offering. Like it just, you, at any sort of reasonable price point, it's just not possible. The, you know, global per capita GDP is not …”
Andreessen: AI training data disputes must be solved by Congress
“In the case of training data, I think that there, I mean, there's a bunch of these copyright, you know, kind of lawsuits happening right now. There's, you know, the big New York Times opening, I won, and there's, you know, been a bunch of others. I think in th…”
Andreessen: Supreme Court will decide if AI chats get Fourth Amendment protection
“On the on the privacy thing. I think that, that one feels like it's a Supreme court thing. To me, it feels like that's the kind of issues that the Supreme court and the, in other words, like whether, for example, your Transcripts are considered your property a…”
Andreessen: The best route into VC is early product development experience
“The best way to do it is to have a track record early as somebody who is like in the loop specifically on new product development. And so somebody who, you know, be like deeply in the trenches at one of these new companies in one of these spaces you know, part…”
Coogan: Knowledge Retrieval Is a $12B Market Set to Grow Significantly
“Knowledge retrieval is clearly a, you know, twelve billion dollar a year market that consumers will pay for. That will probably grow significantly.”
Coogan: AI will see a Cambrian explosion rather than single-company centralization
“So that could be the regime for the next few years as we go into this, like, You know instead of like this centralizing AI force, there's only one company. There's actually like a Cambrian explosion of a ton of companies doing a bunch of different things.”
Sonwalkar: GPT-5 costs half as much as o3
“Also it's half the cost of O three. So it's much cheaper. So it helps you, helps your margins.”
Hays: 85% of ChatGPT's weekly active users are outside the US
“85% of their weekly actives are outside of the U.S.”
Coogan: Model auto-routing gives OpenAI customer leverage and higher margins
“This change is more bullish for the business because it shows that, that OpenAI is a dominant consumer app and they have increasing leverage over the customer to route to cheaper models that will save money and be higher margin.”
Coogan: OpenAI will undoubtedly be able to monetize ChatGPT with ads
“There's no doubt that they'll be able to put ads in this, like the business of the Of the accidental consumer company is as strong as ever”