Show and guests
A conversation on the Pivot podcast, hosted by Kara Swisher and Scott Galloway. Original English title: Why Scott Galloway Sees “Tons of Risk” in Anthropic’s IPO.
Key points
- Asked what worries him, Galloway answers directly: tons of risk, with much more beyond the obvious risk of AI running out of control 24:29
- He calculates Anthropic’s computing commitments at approximately 400 billion dollars; market estimates range from 600 billion to 800 billion dollars. If it cannot sustain 10-fold growth, the numbers do not add up 25:40
- Competitive risks are everywhere: Nvidia could put GPUs directly into phones, and Meta’s free large language model has already topped the app store and reached billions of users 25:54
- Anthropic differs from OpenAI in revenue recognition: Anthropic recognizes the full amount of token sales as revenue, while OpenAI recognizes only a portion 23:48
Background and notes
This article summarizes the discussion of Anthropic IPO risks in this episode of Pivot, covering regulation, computing commitments and competitive threats. The episode also discusses the midterm elections and the Starlink controversy; those sections are not covered here. The subtitles span 0:00–1:01:42 (full duration: 1:01:42); this article covers the portion discussing Anthropic’s IPO. Views and timestamps are paraphrased from the original subtitles, and each timestamp marks the start of the corresponding passage.