AI Hyperscalers and Anthropic

Eric Ries: Why Good Tech Companies Go Bad, and How to Stop It

Topic 8 of 9

Published Jun 1, 2026 · Clip 48:28–51:43 (3:15)

Eric discusses Anthropic, a case study in the book he knows firsthand. He traces the 'why' chain behind Anthropic's lead — superior inference costs, talent, product velocity — all the way back to its unique structure: an ethos of AI safety plus structural protection from outside corruption. He notes doing the right thing doesn't mean being absolutely right; Anthropic gave up a $200 million Pentagon contract and endured bullying to stick to its principles, similar to Costco, and that integrity earns respect even from those who disagree.