Washington University CIO Scott Wilson warns OpenAI and Anthropic face existential threat from cheap Chinese AI models
Scott Wilson, CIO of Washington University’s endowment, argues that OpenAI and Anthropic are overvalued because their massive spending commitments are unsustainable against rising competition from cheaper Chinese AI models. Wilson, who previously profited from an early SpaceX bet, claims that frontier AI labs are not worth their liabilities as open-weight alternatives from China close the performance gap. He cited conversations with portfolio companies that are shifting to these lower-cost models, comparing the situation to high-cost US goods competing with cheap imports. While Wilson’s view contradicts the bullish stance of early OpenAI investor Vinod Khosla, who argues that closed-model infrastructure advantages will lower long-term costs, Wilson’s warning highlights growing concerns about the economic viability of the current AI spending frenzy.
Key points
- Scott Wilson, CIO of Washington University’s endowment, stated that OpenAI and Anthropic are in major trouble due to unsustainable spending commitments.
- Wilson argues that cheaper Chinese models, such as those from DeepSeek and Alibaba, are narrowing the performance gap, making frontier AI labs less valuable relative to their liabilities.
- Wilson noted that many of Washington University’s portfolio companies are switching to open-source models to reduce costs, comparing the trend to US goods competing with low-cost Chinese imports.
- Vinod Khosla, an early OpenAI investor, disagrees with Wilson, arguing that closed-model companies like OpenAI have lower long-term costs due to control over infrastructure and chips.
- Data from OpenRouter shows DeepSeek accounts for 25.3% of text-model requests, compared to 18.6% for OpenAI and 2.9% for Anthropic, supporting Wilson’s argument about the rise of cheaper alternatives.
Background
This debate occurs amid recent concerns about AI safety and regulatory scrutiny, with OpenAI and Anthropic leaders previously urging slower progress to manage risks. Additionally, Anthropic’s upcoming IPO faces scrutiny over its ability to sustain revenue growth amid intensifying competition from OpenAI and open-model providers. The current discussion adds to broader questions about the economic durability of AI companies in the face of rising competition and potential regulatory changes.
Why it matters
The debate over the sustainability of AI spending and the competitive threat from Chinese models could significantly impact the valuations and future strategies of major AI companies. If Wilson’s concerns prove correct, it could lead to a reassessment of the AI industry’s growth trajectory and investment returns, potentially affecting not only OpenAI and Anthropic but also the broader tech sector and its investors.
What to watch
Investors and industry observers will likely monitor the performance of Chinese AI models and the adoption rates of open-weight alternatives in the coming months. The outcome of Anthropic’s IPO and any shifts in OpenAI’s strategy in response to competition will also be key indicators of the industry’s trajectory. Additionally, further developments in AI safety and regulatory discussions may influence the pace of AI development and investment.
Want the full story? Read the original reporting
Read on Business Insider