Chinese open-weight pricing pressure threatens the OpenAI and Anthropic IPO windows

CNBC's read of the Q2 prep work: Chinese models went from roughly 1% of OpenRouter usage in mid-2024 to more than 60% in May 2026, driven by a 5–20× price-per-token gap to closed flagships. That pressure is materially complicating the OpenAI and Anthropic IPO timelines because public-market investors are starting to discount the "closed lab moat" thesis that justified the private-round multiples.

The numbers force the question. Anthropic just priced at $380B post-money on $14B ARR — a ~27× multiple. OpenAI's last private round priced higher. Both numbers assume the closed-lab tier holds its capability premium against an open-weight tier that costs 90% less. If the OpenRouter share data is the leading indicator, the premium is already eroding faster than the IPO docs assume.

The procurement implication for buyers is clean. Q3 2026 RFPs should line-item open-weight inference at the 5-20× price floor alongside the closed-flagship offering. The orgs that build dual-tier inference now lock in the cost line; the orgs that don't will see Q4 budget reviews ask why.

CNBC — cheap AI could derail OpenAI Anthropic IPOs → · Air Street Press — State of AI May 2026 →