Useful context for anyone building on OpenAI's stack: Altman lays out how the company thinks about compute scale and its own silicon, which shapes what capacity and pricing developers can expect downstream.
In this Invest Like the Best episode, Sam Altman recounts OpenAI's history and the origins of ChatGPT and Codex, and discusses the company's approach to hardware and compute, including the 'Jalapeno' chip.
The conversation frames these developments around how abundant, cheap intelligence could reshape the future economy.
Transcript
Sam Altman discusses OpenAI's history, the origin of ChatGPT, Codex, hardware, compute, and the Jalapeno chip, framed around how abundant intelligence could change the future.
podcastWhy humans are AI’s biggest bottleneck (and what’s coming in 2026) | Alexander Embiricos (OpenAI Codex Product Lead)Lenny Rachitsky
podcastOpenAI Codex lead on the new shape of product work | Andrew AmbrosinoLenny Rachitsky
articleImpressions from visiting OpenAI, Anthropic, & CursorGergely Orosz