The All-In Podcast
OpenAI's Identity Crisis, Datacenter Wars, Market Up on Iran News, Mamdani's First Tax, Swalwell Out
Episode Summary
AI-generated · Apr 2026AI-generated summary — may contain inaccuracies. Not a substitute for the full episode or professional advice.
This episode of The All-In Podcast features the 'core four' hosts, joined by special guest Travis Kalanick, co-founder of Uber, to dissect the tumultuous intersection of tech, markets, and politics. The discussion kicks off by scrutinizing New York City's proposed pied-à-terre tax, comparing it to London's detrimental stamp duty, and exploring its potential to destabilize the high-end real estate market while reflecting broader anti-wealth sentiment. The conversation quickly pivots to the high-stakes world of AI, covering OpenAI's internal struggles, Anthropic's rapid growth, and the fierce 'datacenter wars' for essential compute resources.
The panel delves into an internal memo from OpenAI's Chief Revenue Officer, Denise Dresser, which critiques Anthropic's reported $30 billion run rate as inflated and outlines OpenAI's strategic shift toward business customers and agent platforms. They analyze Anthropic's reported 10x year-over-year growth against OpenAI's 3-4x, leading investors to favor Anthropic in secondary markets, challenging OpenAI's $850 billion valuation. A critical point of contention is the 'Fster effect'—the risk of AI labs being throttled by hyperscalers due to compute constraints—highlighting the urgent need for proprietary infrastructure amidst growing public opposition to data center construction across the US.
The discussion further illustrates market irrationality with the bizarre case of 'Allbirds' stock surging 450% after a superficial pivot to 'Newird AI,' reminiscent of the dot-com bubble. Chimath Palihapitiya warns this signifies a 'canary in a coal mine' for massive compute constraints, exacerbated by a shift in public sentiment against AI and data centers, seen as 'temples of the wealthy.' Saxs elaborates on the multi-faceted reasons for data center unpopularity, from fears of increased electrical rates to strategic 'doomerism' lobbied by certain groups, including Anthropic, potentially now backfiring as they seek their own infrastructure.
The conversation also touches on political maneuvering, exemplified by Eric Swalwell's withdrawal from the California governor's race amidst coordinated allegations, prompting speculation about behind-the-scenes party influence. Finally, the panel grapples with the paradox of a market hitting all-time highs despite geopolitical tensions and high valuation indicators like the Schiller PE and Buffett Index. Travis Kalanick introduces 'Trump's weather vane' theory, suggesting the stock market dictates the former president's policy, while Jason Calacanis optimistically posits that AI's exponential efficiency gains will eventually justify these valuations, even as Chimath remains risk-off due to a lack of scaled, proven profits in large enterprises.
Listeners will gain a comprehensive understanding of the complex forces shaping the current tech and investment landscape, from the disruptive potential of AI and its infrastructure challenges to the subtle yet powerful influence of political dynamics and public sentiment on market trends.
👤 Who Should Listen
- Investors tracking the AI industry and the competitive landscape between frontier models like OpenAI and Anthropic.
- Real estate developers and investors concerned about the impact of wealth taxes and regulatory shifts on property markets in major cities.
- Tech executives and founders navigating the challenges of AI adoption, change management, and securing compute infrastructure.
- Policymakers and community organizers interested in the public perception of AI, data center development, and the populist backlash against technological progress.
- Anyone interested in the intersection of politics, technology, and finance, especially concerning market reactions to geopolitical events and political maneuvering.
🔑 Key Takeaways
- 1.Mamdani's proposed 3.9% pied-à-terre tax in New York for homes over $5 million is predicted to cause a 'massive impact on demand for second homes,' potentially crashing the high-end market. [00:46, 01:10]
- 2.OpenAI's internal memo called Anthropic's $30 billion run rate 'cap' and inflated by revenue share accounting, signaling OpenAI's pivot to focus heavily on business customers and agent platforms. [11:14]
- 3.Anthropic's reported 10x year-over-year revenue growth far outpaces OpenAI's 3-4x, leading secondary markets to value Anthropic higher due to its ability to fund growth through revenue rather than just investment. [22:38, 19:19]
- 4.The market for AI compute is 'massively constrained' by power, land approvals, and growing public opposition to data centers, posing a 'five alarm fire' for frontier AI labs like Anthropic and OpenAI. [39:42, 41:46]
- 5.The 'Allbirds' stock surge of 450% after pivoting to 'Newird AI' exemplifies 'peak bubble behavior' and the irrational market reaction to superficial AI branding. [33:35, 38:41]
- 6.Public sentiment in America is 'incrementally getting more and more negative on the whole subject matter rit large' regarding AI, leading to local opposition and voting down data center projects. [40:43]
- 7.Political maneuvering, such as the coordinated release of allegations against Eric Swalwell to force his withdrawal from a governor's race, highlights the power of 'insiders' to shape political outcomes. [61:14, 62:14]
- 8.The stock market is exhibiting 'weird dispersion,' with a few companies hitting all-time highs while traditional valuation indicators like the Schiller PE and Buffett Index suggest overall high market risk. [70:25]
💡 Key Concepts Explained
Pied-à-Terre Tax
A proposed New York City tax of 3.9% on second homes valued over $5 million annually. The episode discusses how this tax could significantly reduce demand for non-primary residences, potentially crashing the high-end real estate market and deterring investment, similar to London's stamp tax. [00:46, 01:02]
Datacenter Wars
The intense strategic competition among AI labs and tech giants to build and control massive compute infrastructure (GPUs, power, and land). This battle is crucial for scaling AI models and is complicated by resource scarcity and increasing public opposition to new data center construction. [27:32, 39:42]
Fster Effect
A phenomenon where a promising technology or platform, despite early success, fails to scale effectively due to infrastructure limitations (like compute capacity), allowing competitors with better scaling capabilities to overtake it. This is used to explain how throttling compute access can kneecap frontier AI labs. [28:33, 41:46]
Trump's Weather Vane Theory
Travis Kalanick's theory that Donald Trump's political actions and policy decisions are heavily influenced by the stock market's performance, which he uses as a 'weather vane.' This suggests Trump aims to keep the market trending upward, even if it involves making initial controversial moves and then course-correcting. [73:31]
Market Dispersion
A confounding market condition where traditional valuation indicators (e.g., Schiller PE, Buffett Index) suggest overall market highs and potential risk, yet a small number of mega-cap companies are achieving unprecedented growth and all-time high valuations. This creates a challenging environment where data can be selectively used to support various investment biases. [70:25]
⚡ Actionable Takeaways
- →When evaluating real estate investments in major cities, consider the increasing risk of wealth taxes like the proposed pied-à-terre tax, which can significantly impact demand and property values. [00:46, 07:09]
- →For AI companies, prioritize securing proprietary compute infrastructure (land, power, data centers) to avoid reliance on hyperscalers and mitigate the risk of throttled access and capacity. [27:32, 41:46]
- →Be cautious of companies making superficial pivots to 'AI' solely for stock market pumps, recognizing this as a potential indicator of speculative bubble behavior. [33:35, 38:41]
- →When implementing AI in large organizations, acknowledge that 'change management' is a significant human challenge, requiring careful navigation of existing processes and overcoming resistance from middle managers and bureaucrats. [83:42]
- →For new data center projects, explore solutions for onsite power generation, such as those offered by Bloom Energy, to expedite clean air permits and reduce dependency on local grids. [39:42]
- →Entrepreneurs in the AI space should prioritize developing consumer-facing applications that clearly demonstrate positive value to the average person to counter growing public skepticism and 'doomerism.' [46:52]
- →Investors should adopt a 'risk-off' posture during periods of high market valuation metrics, even if specific sectors or companies are performing exceptionally well, and wait for clearer opportunities or deleveraging events like IPOs. [82:42, 75:35]
⏱ Timeline Breakdown
💬 Notable Quotes
“"It's a dog whistle to say that's the next United Healthcare CEO and in the week that a fire a Molotov cocktail and a bullet gets shot into Sam Alman's house, it's deadly serious." [02:30]”
“"If Anthropic is funding theirs through revenue and other folks are funding it through investment, there's like a short-term that's a short-term solve, but it the long run is whoever is scaling their actual usage and system and ultimately with contribution profit that then soaks up the need for investment. That's a that's a that's a very scary machine if you're competing against it." [19:19]”
“"The reason is because underneath at the core of it all is there's a tide that is shifting on AI. the American population is incrementally getting more and more negative on the whole subject matter rit large..." [40:43]”
“"The stock market is Trump's weather vein and he will go to the place that makes the stock market come up and if he's go and and if it's up and it's too high, he almost feels like it's too easy. So he makes it hard on himself again and then he brings it back up and the traitors are getting used to it I think." [74:34]”
“"The efficiency boom that we're going to see at these companies, whether it's Meta or Uber or Airbnb, whoever executes it properly, is going to be phenomenal. The earnings will be insane." [78:20]”
“"The agents are what you you have to be human in the loop with the agent. The agent has no taste. The agent can do repetitive tasks. the agent is not going to do something novel and they quickly can get lost in the forest." [85:45]”
Listen to Full Episode
📬 Get weekly summaries like this one
No spam. Unsubscribe anytime. By subscribing you agree to our Privacy Policy.
Continue Exploring
