US Businesses Could Face $12 Billion in Higher Costs Annually Under Potential Chinese AI Model Ban: Report
Complete. Here is the key summaryA potential U.S. ban on Chinese open-weight AI models could cost American businesses $3 billion to $12 billion annually, according to Georgia Tech researcher Daniel Yue. Major tech firms like Nvidia and Meta oppose the ban, warning it harms competition and innovation. Meanwhile, OpenAI slashed prices amid falling inference costs driven by cheaper Chinese alternatives. Experts argue targeted chip export controls are more effective than model bans.
A potential U.S. ban on Chinese open-weight AI models could come at a massive cost to American businesses, as companies increasingly rely on lower-cost Chinese AI solutions, according to estimates from a researcher.
Increase in Costs
Increased costs from restrictions on Chinese open-weight AI models could add $3 billion to $12 billion to the broader U.S. economy, depending on adoption levels, according to Georgia Tech assistant professor Daniel Yue, as reported by the South China Morning Post.
His estimate used token usage and pricing differences between open- and closed-source AI models from July 21–27, based on usage data from AI platform OpenRouter. Yue emphasized these are rough, order-of-magnitude estimates, not precise projections, due to the difficulty of measuring decentralized usage
Yue said the economic impact is still uncertain because it’s unclear whether U.S. companies would adopt closed AI models or stop using some AI workflows altogether.
Foundation Capital partner Jaya Gupta warned that banning open-weight AI models could sharply reduce AI demand, risking a collapse in the debt-funded AI infrastructure market as data center investments depend on sustained demand.
Read Also: Meta CEO Mark Zuckerberg Says Chinese AI Model Ban Not an 'Effective Solution', Calls for Fewer Barriers: 'There's…This Question of Regulatory Capture'
Moonshot AI Fuels US Concerns
Washington’s concerns over Chinese open-weight AI models intensified after Beijing-based Moonshot AI launched Kimi K3 in late July. The model rivaled leading offerings from OpenAI and Anthropic on some benchmarks, prompting renewed Trump administration efforts to restrict foreign open-source AI models amid allegations that Moonshot AI infringed U.S. intellectual property.
Major U.S. tech firms, including Nvidia Corp. (NASDAQ:NVDA), Palantir Technologies Inc. (NASDAQ:PLTR), and Meta Platforms Inc. (NASDAQ:META), urged the government not to restrict open-weight AI models, warning that premature regulations could hurt competition and push innovation overseas.
OpenAI Slashes GPT-5.6 Prices
OpenAI CEO Sam Altman, on Friday, announced an 80% price cut for the lightweight GPT-5.6 Luna model and a 20% reduction for the mid-tier GPT-5.6 Terra. The move comes as global spending on large language model inference fell from $2.07 to $1.67 per million tokens between early June and early July, a decline Goldman Sachs attributed in part to growing adoption of lower-cost Chinese open-weight AI models.
Meanwhile, Anthropic CEO Dario Amodei said he has never supported banning open-weight AI models, rejecting claims that Anthropic seeks to shield its closed-model business from open-source rivals. He argued that restricting Chinese open-source models would do little to stop bad actors and instead proposed targeted export controls on advanced AI chips and chipmaking equipment to China and other authoritarian governments, calling them the most effective way to limit China’s AI capabilities.
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
Read Also: Steve Eisman Says He'd Be 'Petrified' Running OpenAI or Anthropic as Cheap Chinese Models Threaten Price War
Photo courtesy: Shutterstock