The discussion
The hosts are broadly bullish on Anthropic, with near-universal agreement that Claude — particularly its coding capabilities — represents best-in-class product quality, and that the company's revenue trajectory (reportedly ~10x annual growth, EBIT-positive, and potentially $80–100B ARR by end of 2026) is historically unprecedented; Sacks, Chamath, Friedberg, and guests Gerstner, Baker, and Benioff all cite these metrics as core to their conviction. There is also broad agreement that Anthropic's enterprise focus, MCP standard adoption, and coding-agent specialization have created durable competitive advantages, though Chamath and Sacks both flag cost (Anthropic being expensive relative to open-source alternatives) and compute constraints as near-term tactical risks rather than structural flaws. The most pointed disagreement centers on Anthropic's safety and regulatory posture: Sacks, Chamath, and guest Gurley are sharply critical, arguing its "safe AI" branding and lobbying amount to calculated regulatory capture designed to entrench monopolistic control, while the bullish financial theses from Gerstner and Friedberg largely set this concern aside. Chamath holds the broadest range of views across time — early bearishness on capital constraints (2024), evolving to high-conviction bullishness on enterprise dominance, but punctuated by ongoing skepticism about the durability of enterprise revenue, the lack of owned compute infrastructure, and the sincerity of its safety narrative.