CNBC tech reporter Deirdre Bosa argues that cheap Chinese AI models could derail OpenAI and Anthropic's expected IPO valuations of over $800B each, since both valuations assume the labs will hold their market share and pricing power. The thesis hinges on a widening cost gap: AI benchmarking firm Artificial Analysis found Anthropic's Claude costs $4,811 to run through 10 standard evaluations, compared to DeepSeek at $1,071 and Zhipu's GLM at $544. Adoption is already shifting, with Chinese models growing from roughly 1% of usage on AI marketplace OpenRouter in 2024 to more than 60% in May 2026. Anthropic itself acknowledged the pressure in a May policy paper, saying U.S. models are only “several months ahead” of Chinese ones and warning that Beijing is “winning in global adoption on cost.” OpenAI's confidential IPO filing is expected as soon as this week.
