Keen On America

Why Even Sam Altman Wants to be Gary Marcus: From Son of Sam to Son of Gary in a single ChatGPT Release

Aug 27, 2025
Gary Marcus, a notable AI skeptic and author of 'Taming Silicon Valley', shares his journey critiquing the hype around AGI and large language models. He reveals how the underwhelming launch of GPT-5 shifted the narrative from exponential growth to cautious realism, catching even Sam Altman’s attention. Marcus highlights OpenAI’s unsustainable financial model, the limitations of current AI technologies, and the urgent need for diverse approaches in AI development. His contrarian views are now gaining unexpected support in the tech industry.
Ask episode
AI Snips
Chapters
Books
Transcript
Episode notes
INSIGHT

Altman Adopts A More Cautious Tone

  • Gary Marcus says Sam Altman is not becoming like him; rather Altman is starting to adopt Marcus's cautionary language.
  • Marcus argues OpenAI is downscaling AGI claims after overhyping GPT-5.
INSIGHT

GPT-5 Shifted The Narrative

  • GPT-5 delivered incremental improvements, not a 'quantum leap' after 34 months of work.
  • That failure shifted the narrative from exponential progress to diminishing returns in AI scaling.
INSIGHT

OpenAI's Cash-Runway Problem

  • Marcus claims OpenAI faces a cash-flow problem despite huge valuations and record revenue.
  • He estimates runway at roughly six to eighteen months, forcing continual fundraising and hype.
Get the Snipd Podcast app to discover more snips from this episode
Get the app