Could China’s Cheap AI Models ‘Crash’ The US Stock Market In 2026? CEO Issues Stark Warning For OpenAI And Anthropic: ‘The Bubble Is…. About
Polsia AI CEO Ben Cera ‘s argued that China is releasing state-of-the-art open-source (SOTA) AI models in the U.S., like GLM-5.2, that perform nearly as well as leading frontier models at a fraction of the cost, making it hard for companies to justify spending $100 million when similar capabilities may be available for around $5 million. In an X post on Tuesday, he cautioned that this would force enterprises to slash AI spending from millions to fractions, collapsing OpenAI and Anthropic valuations. “China will crash the US stock market this year,” he wrote, while adding, “The bubble isn’t AI. It’s the price of AI. And it’s about to pop.” In a subsequent reply to another user, Cera acknowledged that the “AI Model only” companies may be the biggest losers while hyperscalers like Alphabet Inc. (NASDAQ: GOOGL ) (NASDAQ: GOOG ) w...
2, that perform nearly as well as leading frontier models at a fraction of the cost, making it hard for companies to justify spending $100 million when similar capabilities may be available for around $5 million. In an X post on Tuesday, he cautioned that this would force enterprises to slash AI spending from millions to fractions, collapsing OpenAI and Anthropic valuations. “China will crash the US stock market this year,” he wrote, while adding, “The bubble isn’t AI. It’s the price of AI.
” In a subsequent reply to another user, Cera acknowledged that the “AI Model only” companies may be the biggest losers while hyperscalers like Alphabet Inc. (NASDAQ: GOOGL ) (NASDAQ: GOOG ) would remain unscathed. “Google has an infinite war chest. they’ll be fine,” he stated.
China will crash the US stock market this year. They're dumping SOTA open-source models on the US, nearly as good as the frontier labs, basically free. Nobody can refuse cutting a $100M AI bill down to $5M. S.
enterprises are going to increasingly favor Chinese open-source models over frontier labs like OpenAI or Anthropic. S. S. S.
2 model and its strategy of building sovereign large language model systems. 5 are some of the other open-source models. 8 on a key agentic AI benchmark at roughly one-fifth the cost. The model has seen rapid developer adoption and is particularly strong at enterprise-focused tasks such as planning, coding, testing, and automation, unlike DeepSeek, which many viewed as a one-off chatbot breakthrough, reported CNBC.
S. technology. S. Commerce Department’s Entity List in January 2025 over alleged support for China’s military AI efforts, OpenAI said that the company made significant global progress by providing AI infrastructure to governments.
S. 2. 6, enterprises are increasingly weighing models that are less vulnerable to policy-driven access restrictions. Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published editors.
Image via Shutterstock Read Also: Box CEO Aaron Levie Says AI Race Hinges On Whether 'Open Weights AI' Stays 'A Close Second'