China’s President Xi Jinping used the opening of Shanghai’s World Artificial Intelligence Conference on July 17 to call for international AI cooperation and a greater Chinese role in global AI governance. The remarks coincide with a fast-moving commercial challenge to US frontier-model vendors, including Anthropic, but they do not establish a specific Chinese plan to “cripple” the company.
As reported by the Associated Press, Xi said AI development should be a global effort rather than “a solo performance” by one country. He also criticized restrictions on technology sharing framed as national-security measures. China’s foreign ministry said the July 17–20 conference is intended to build international consensus on AI development and governance.
The more immediate issue is Moonshot AI’s Kimi K3, released around the conference. According to Axios, the Beijing-based company says the model performed strongly in public coding and general text rankings against leading Anthropic and OpenAI systems, while carrying substantially lower costs. Moonshot’s own site lists Kimi K3 as its latest model, though vendor benchmark claims should be treated as marketing until independently reproduced across real workloads.
The timing matters because Anthropic has formally begun IPO preparations. Anthropic announced on June 1 that it had confidentially submitted a draft registration statement to the US Securities and Exchange Commission for a proposed initial public offering. The company’s May funding round reportedly valued it at $965 billion, with annualized revenue of $47 billion.
That puts a heavy premium on continued technical leadership and the ability to defend its pricing.
Moonshot’s K3 release does not prove that allegation, and China’s government rhetoric at WAIC did not address Anthropic by name. But the episode shows why American AI firms view low-cost Chinese competitors as both a pricing threat and a model-security risk.
Microsoft is already publicly focused on the cost angle. Microsoft AI chief Mustafa Suleyman told Bloomberg in June that Anthropic was “extremely expensive” and that customers were seeking alternatives, as reported by CIO. Microsoft has obvious incentives to steer users toward its own AI products, so that assessment is not neutral. Still, it reflects a wider enterprise concern: agentic coding and workflow tools can generate large token bills quickly.
As reported by the Associated Press, Xi said AI development should be a global effort rather than “a solo performance” by one country. He also criticized restrictions on technology sharing framed as national-security measures. China’s foreign ministry said the July 17–20 conference is intended to build international consensus on AI development and governance.
A sharper competitive problem for Anthropic
The more immediate issue is Moonshot AI’s Kimi K3, released around the conference. According to Axios, the Beijing-based company says the model performed strongly in public coding and general text rankings against leading Anthropic and OpenAI systems, while carrying substantially lower costs. Moonshot’s own site lists Kimi K3 as its latest model, though vendor benchmark claims should be treated as marketing until independently reproduced across real workloads.The timing matters because Anthropic has formally begun IPO preparations. Anthropic announced on June 1 that it had confidentially submitted a draft registration statement to the US Securities and Exchange Commission for a proposed initial public offering. The company’s May funding round reportedly valued it at $965 billion, with annualized revenue of $47 billion.
That puts a heavy premium on continued technical leadership and the ability to defend its pricing.
The distillation dispute is the other half of the story
There is a direct conflict between Anthropic and Moonshot that goes beyond ordinary product competition. In February, Anthropic said it had detected what it described as industrial-scale efforts by DeepSeek, Moonshot and MiniMax to extract Claude’s capabilities through “distillation” — using outputs from a stronger model to train another system. Anthropic said it does not offer commercial Claude access in China or to overseas subsidiaries of Chinese companies for national-security reasons.Moonshot’s K3 release does not prove that allegation, and China’s government rhetoric at WAIC did not address Anthropic by name. But the episode shows why American AI firms view low-cost Chinese competitors as both a pricing threat and a model-security risk.
Microsoft is already publicly focused on the cost angle. Microsoft AI chief Mustafa Suleyman told Bloomberg in June that Anthropic was “extremely expensive” and that customers were seeking alternatives, as reported by CIO. Microsoft has obvious incentives to steer users toward its own AI products, so that assessment is not neutral. Still, it reflects a wider enterprise concern: agentic coding and workflow tools can generate large token bills quickly.
What Windows and IT teams should do
For Windows developers and administrators, this is not a reason to swap providers based on a single leaderboard. It is a reason to tighten AI procurement:- Test candidate models against representative code, documents and support workflows.
- Compare total token, hosting and integration costs rather than headline API prices.
- Check data handling, regional hosting, support, export-control and contractual requirements before using Chinese-hosted services or open weights.
- Keep provider abstraction in place where possible, particularly for Copilot-adjacent internal tools.
References
- Primary source: 24/7 Wall St.
Published: 2026-07-17T13:06:03+00:00
Loading…
247wallst.com - Related coverage: axios.com
China just erased America's AI lead
Kimi's K3 model has caught up to the American frontier.www.axios.com