TBPN Podcast Deep Dive: “AI Boom Stalls, Blue Owl’s Next Move, ChatGPT Growth Concerns”

Date: November 13, 2025
Hosts: John Coogan & Jordi Hays
Special Guests:
Jeffrey Katzenberg, The Winklevoss Twins, Andrew Dudum (Hims & Hers), Spencer Peterson (Coatue), Max Hodak (Science), Melissa Tokmak (Netic), Tomás Puig (Alembic)

Episode Overview

This episode revolves around major shifts in the AI and tech finance landscape: the faltering pace of AI’s commercial growth, the new prevalence of debt in tech infrastructure (with Blue Owl’s bold moves in data center financing), skepticism around ChatGPT’s expansion, and the evolving business models in health, biotech, and foundational AI.

The format features an initial deep-dive by the hosts, followed by a marathon of high-profile guests discussing everything from AI investing, private credit, neurotech, practical applications in health, and the real economics behind digital business.

Key Discussion Points

1. Slowing AI Growth & ChatGPT's Plateau

[00:11-05:00, 25:50-38:00]

• OpenAI’s ChatGPT has hit “deceleration, not decline,” per Sarah Fryer (CFO) and industry signals.
  • “It’s not in decline, it’s maybe deceleration.” — John [00:11]
  • Possible causes: saturation, content restriction, competition.
  • Discussion of external signals from app/store analytics, especially in Europe and the back-to-school period.
  • Unlocking “erotica content” acknowledged for future user growth stimulus.
  • Quote: “I don’t think them announcing they’re getting into erotica is a sign of strength.” — Jordi [29:18]
• Industry-wide implications: Is this just a phase on the way to normal software product adoption curves, or a warning that the AI infrastructure wave (with immense capex and debt) may precede sustainable demand?

2. The Rise of Debt in Tech – Blue Owl’s 'Hyperion' Data Center Deal

[05:00-25:00+]

• Debt is now 'the air' in AI infrastructure.
  • The $1.4T in OpenAI’s planned spending is “the same size as the oil and gas market.” — John [17:29]
  • Blue Owl Capital’s role: financing Meta’s $27B “Hyperion” data center, providing $3B upfront, with Meta as a long-term tenant.
  • Strategic interplay between private credit (Blue Owl, Ares) and traditional banks (e.g., Jamie Dimon’s “cockroach” warning).
• Risk/Reward mechanics:
  • Data centers’ worth is hard to underwrite (uncertain GPU lifespans vs. predictable oil fields).
  • Blue Owl’s BDC structure: yields for shareholders, high leverage, big diversification.

Quote:

• “Private Asset Star. Blue Owl has been flying high. Is it too close to the sun?” — Barron’s (read aloud) [12:53]
• “Blue Owl is the pretty girl at the dance right now.” — David Williams (Wall Street) [18:34]

Notable Segment:
Explainer on Blue Owl’s BDC structure and why tech is looking more like “oil and gas”:
[20:05-22:46]

3. Macro Market Jitters & Tech Downturn

[25:24-54:00]

• AI infrastructure stocks and suppliers are down:
  • Nvidia (-4%), CoreWeave (-8%), broader uncertainty in the “early debt cycle,” and possible signs of bubble bursting.
• Comparisons to historical tech cycles:
  • OpenAI’s high spending vs. early Google’s “infinite money glitch.”
  • Application of debt in hyperscaler projects; skepticism about current revenue models keeping up with capex.

Quote: “There’s far more AI computing infrastructure spending than there is AI revenue, a gulf widening by the day.” — John [53:01]

4. Guest Interview: Spencer Peterson (Coatue) on AI’s Investment Structure

[59:36-80:11]

• Cursor’s breakout success: Fastest SaaS to $1B in revenue; “transcendant” company in the code/AI dev productivity space.
  • Coatue, a major investor, also backs OpenAI and Anthropic—argues the future is “not winner take all.”
  • Market is a “blue ocean,” B2B demand can support multiple giants.
• Defensive strategies, public/private market disconnects, and founder tunnel vision:
  • “Tunnel vision is everything… Don’t get distracted by the markets.” – [73:01]
  • Most top AI companies are “awash in cash.” Jitters are more about optics and “anxiety displacement.”
• “Anxiety displacement” as new jargon for market panic:
  • “People are definitely reading too much into anything and everything sometimes.” — Spencer [77:44]

5. Winklevoss Twins Announce Cypherpunk ZK-Cash Trust (DAT)

[86:11-88:38]

• A new privacy-focused, ZK-based digital asset trust:
  • Structure: takeover/rebranding of an existing biotech firm as publicly-traded ZK-Cash vehicle.
  • Long-term holding focus—large insider ownership, low “fast money” flow.
  • Noted momentum in ZCash and wider “DAT” sector, but overall market choppiness acknowledged.

6. Max Hodak (Science, Former Neuralink) on Brain Interfaces & AI

[90:04-122:15]

• Science’s dual track: pragmatic retinal prosthesis, and “biohybrid” cortex interfaces.
  • Restoring sight to the blind using “solar-powered” chips, vision much improved vs. existing approaches.
    • “We think a visual prosthesis is a brain-computer interface; so is a cochlear implant.” — Max Hodak [95:12]
  • Implant is “outpatient, nearly as simple as Lasik,” powered by projected infrared image from glasses.
• Long-term vision: “Biohybrid” interface grows new biological connections with the brain—massively higher bandwidth, no drilling.
• AI cross-pollination:
  • “At the very beginning…OpenAI and Neuralink were in the same building… Who is going to learn more from whom?”
  • “Turns out AI learned the concept of a neuron and that’s basically it.” — Hodak [102:34]
• On ethics of “the merge,” transhumanism and agency:
  • “I don’t like ‘transhumanism’…what we’re doing should make you just as much you as you’ve ever been.” [118:27]
• AI’s leverage in biotech:
  • Used for regulatory, document review, and—crucially—AI-driven protein engineering for optogenetics.

7. Andrew Dudum (Hims & Hers) on Health, Diagnostics, and Compounding

[122:17-140:56]

• Hims & Hers: biggest digital telehealth platform in the US.
  • Eight years old, “millions of patients, 10–15k/day digital visits,” now $2B+ revenue, profitable, rapid growth.
  • New Labs initiative: 120+ biomarkers for $199–$499, soon aiming for $30—pricing labs as potential loss leader/lead-gen.
• Strategic vision:
  • Labs should be “free”—value is in the data, customization, and long-term relationship.
  • “Anybody in the labs business today whose core business is labs needs a new business.” — Dudum [126:43]
• GLP-1 drugs (weight loss/diabetes):
  • Prices falling from $1500/mo to $150, predicting $50 within years as new entrants/generics expand.
• On compounding & regulatory battles:
  • “We play very clearly by the rules…this is the first time in history for this level of personalization at scale.” [133:27]
• Candid on volatility:
  • “The team has a stomach of steel…It’s not glitz and glamour, it’s heads-down, big opportunity thinking.” [135:01]

8. Melissa Tokmak (Netic) on AI for Essential Services

[141:20-154:07]

• Netic builds AI-powered “revenue engines” for essential businesses (HVAC, plumbing, electrical, etc.)
  • “Backbone of the American economy”—companies that are countercyclical, always needed.
  • AI agents help with surge handling (disasters, weather), customer communications, prediction and repeat engagement.
• $23M Series B at $450M cap led by Founders Fund (4x step up)
  • Low dilution, strong fundamentals, public recognition that “nimble teams” are a hiring magnet.
• PE rollups as key customers; AI ROI is the main sales pitch (not “strategy” slides).

9. Tomas Puig & Jeffrey Katzenberg (Alembic) – Causal AI for Attribution

[156:02-177:52]

• Alembic, $145M round, builds causal AI to answer the oldest question in marketing: “What worked?”
  • Goes beyond “last click” or standard attribution by ingesting huge amounts of unstructured/semi-structured data (“a trillion connections”).
  • Proven use cases: Delta’s Olympics sponsorship, brand/influencer effectiveness.
  • “It’s like being in a pitch-black room and you’d rather have a really good guess at where the exit is than wander around.” — Tomas [167:27]
• Accenture: from client to largest-ever venture investor, serves as both validator and GTM channel.
• Long-term vision: democratize these insights for much smaller brands, so “the best products win, not just the ones with resources for bespoke analytics.”
• Validation & backtesting: core to product credibility; both synthetic and real-world (business, fMRI) datasets used.
• Fun moment: hosts and Tomas discuss their desire to run causal AI on podcast billboards to finally know if they “worked.”

Notable Quotes & Memorable Moments (with Timestamps)

• “It’s not in decline, it’s maybe deceleration.” — John [00:11]
• “Blue Owl is the pretty girl at the dance right now.” — Wall Street trader [18:34]
• “Data is the new oil? Maybe tech is about to look like oil and gas—lots of capex, lots of debt.” — John [04:12]
• “Private companies need tunnel vision… don’t get distracted by markets, just build.” — Spencer Peterson [73:01]
• “People are definitely reading too much into anything and everything sometimes…anxiety displacement.” — Spencer [77:44]
• “As far as I’m concerned, the body exists to keep the brain alive — I’ll be annoyed if my pancreas kills me.” — Max Hodak [115:43]
• “It’s not glitz and glamour, it’s a heads-down team. A stomach of steel.” — Andrew Dudum [135:01]
• “Causal AI is about being directionally correct, not specifically wrong.” — Tomas Puig [167:13]

Timestamps of Key Segments

• ChatGPT Growth Concerns: [00:11-06:40, 25:50-38:00, 77:39-80:11]
• Blue Owl/Hyperion Data Center Deep Dive: [09:20-22:00]
• Market Jitters & The Debt Cycle: [25:24-54:00]
• Spencer Peterson (Coatue, Cursor) Interview: [59:36-80:11]
• Winklevoss - Cypherpunk Announcement: [86:11-88:38]
• Max Hodak (Science) Interview: [90:04-122:15]
• Andrew Dudum (Hims & Hers) Interview: [122:17-140:56]
• Melissa Tokmak (Netic) Interview: [141:20-154:07]
• Alembic, Tomas Puig & Jeffrey Katzenberg Interview: [156:02-177:52]

Thematic Highlights & Host Tone

The episode balances alarm (“is the boom over? is the market in trouble?”) with irreverence, deep technical insight, and a fixation on actionable economics. Guests are pressed for realities rather than fluff. Market pain is a recurring theme, leavened by playful banter and meme references.

Memorable banter includes the running “need a guy for everything” AI metaphor, the jest about OpenAI’s next business model being “running a lawn mowing service,” and exasperation over counterfeit TVPN merch.

Closing Note

This was a “capital markets” episode at its core: semiserious, highly analytical, and at times sardonic—a must-listen for tech/AI investors, founders, and observers who want a full map, not just the headlines, of where the AI-fueled future is actually going.