Ep. 4: Bill Gurley: Don't Protect Anthropic. Also, Let Washington Fight Open With Open.

Bill Gurley spent the week accusing Anthropic of seeking government protection, then praised Washington for understanding that the answer to open-source AI is more open AI. Alex and Priya decode why that tension fits Benchmark's open-infrastructure incentives, especially after Cerebras turned into a multi-billion-dollar win.

VC Hot Take of the Week
VC Hot Take of the Week@Soren K
Ep. 4: Bill Gurley: Don't Protect Anthropic. Also, Let Washington Fight Open With Open.
0:0010:55

Episode guide

This week, Alex and Priya decode Bill Gurley's anti-Anthropic broadside: the claim that a frontier lab asking Washington for help is really trying to protect pricing power. The twist is that Gurley also praised Washington for understanding that the way to fight open-source AI is with more open AI, a stance that lines up neatly with Benchmark's multi-billion-dollar Cerebras exposure.

Show notes

  • Gurley's loudest post framed Anthropic's June Senate letter as a competitive move: if customers are finding cheaper alternatives, do not ask Washington to protect the incumbent.
  • The second move was not simply anti-government. Gurley later endorsed the idea that open models are a strategic weapon, not just a cost-saving trick.
  • The incentive layer runs through Benchmark's Cerebras stake. Cerebras benefits from faster inference demand, model pluralism, and enterprise pressure to buy more intelligence per dollar.
  • The episode treats the portfolio read as incentive alignment, not as proof of coordination.

Statement Timeline

  1. June 24, 2026, 21:18 UTC — CNBC reported that Anthropic had sent a June 10 letter to the U.S. Senate Banking Committee accusing operators affiliated with Alibaba and Qwen of a large-scale distillation campaign.
  2. June 26, 2026, 08:42 UTC — Gurley posted that Anthropic was "aggressively" begging for government protection because customers were finding cheaper alternatives; the post showed 921,809 views in the detail record captured for this episode.
  3. June 26, 2026, 12:41 UTC — Gurley followed with the "Why write letters to DC? Just use AGI" post; the detail record showed 508,640 views, 459 reposts, and 107 quote posts.
  4. June 26, 2026, 22:08 UTC — CNBC published its open-source AI market context, reporting that Zhipu's GLM 5.2 was near Anthropic's Opus 4.8 on one agentic benchmark at roughly one-fifth of the cost.
  5. June 30, 2026, 12:46 UTC — Gurley argued against putting a high fence around the United States and said there was no need to protect "the fastest growing company of all time."
  6. July 1, 2026, 01:45 UTC — Gurley wrote that the only way to fight "open" is with "open," praising a Washington voice for understanding the stakes.
  7. Contradiction flagged in this episode — The posts were separate statements over several days, not a single statement. The tension is that Gurley rejects government protection for a closed lab while welcoming Washington's strategic support for open-model competition, a position that also supports Benchmark's infrastructure narrative.

Audio note

The episode runs about 10 minutes and 55 seconds, with an 8-second intro and 10-second outro using the channel theme.

Sources

Related content

  • Sign in to comment.
More from this channel
View the full content archive of "VC Hot Take of the Week"Explore more channels on Discover