Calls for federal AI regulation in the United States and a possible global slowdown in the development of frontier models are creating uncertainty for enterprises. Companies are weighing whether future restrictions could affect their use of open-weight models, particularly those developed in China, as demand grows for lower-cost alternatives to proprietary systems.

On Monday, Chinese Foreign Ministry spokesperson Guo Jiakun described recent calls for an AI slowdown from frontier AI companies Anthropic and OpenAI as “fear mongering, confrontation and vicious competition.”

Jiakun’s comments followed an essay published over the weekend by Anthropic CEO and co-founder Dario Amodei. In the essay, Amodei called for the global AI industry to slow the advancement of frontier-model capabilities and urged democratic countries to coordinate with frontier AI companies.

Amodei’s essay came four days after OpenAI’s chief global affairs officer published a statement calling for a federal AI policy. The statement was released on the same day that an AI researcher who had worked for Anthropic resigned, saying that neither OpenAI nor Anthropic acts responsibly. The researcher also claimed that the technology could lead to human extinction.

These developments followed a July appeal by employees at Anthropic, Google, Meta and OpenAI for the U.S. government to support international governance intended to pace AI development.

Enterprises on uncertain ground

As calls for government action increase, enterprises must make technology decisions amid additional uncertainty surrounding a field that is already changing rapidly.

“Enterprises are having to choose AI platforms, vendors, skills, architectures and operating models while knowing that the underlying technology may change before those investments fully mature,” said Kashyap Kompella, founder of RPA2AI Research. “The showdown debate adds several new uncertainties at once.”

Among the questions facing companies are whether frontier-model development will continue at its current pace or slow, whether open models could become subject to restrictions, and whether some important AI capabilities will remain under private control. Those questions also make vendor selection, model access and governance requirements part of the same technology-planning decision rather than separate issues.

David Nicholson, an analyst at Futurum Group, said some of these concerns could serve the interests of OpenAI and Anthropic. Enterprises are increasingly recognizing that, despite the performance of frontier models, those systems cost more than open models. In some cases, companies can accomplish more at lower cost by using smaller, open-weight or fully open-source AI models, he said.

“Every day that goes by, fewer and fewer workloads legitimately need what the leading edge of frontier models can deliver,” Nicholson said, noting that Anthropic and OpenAI are approaching highly anticipated and potentially lucrative initial public offerings.

“There is no question that the leading edge of frontier models is amazing, but when you start pricing the cost of what you’re paying for and looking at the task at hand, there are more economically efficient ways to do what OpenAI and Anthropic are both offering to the market,” he added.

Pressure from open models

With OpenAI and Anthropic facing increasing price pressure from developers of mostly open models, some analysts see their calls for federal AI regulation as being connected not only to broader safety concerns but also to efforts to maintain their position in the AI market.

“In the interest of preserving their duopoly in this space, they are willing to let the government regulate them,” Nicholson said. “What they’re asking for is for the federal government to step in and protect them from Chinese competition.”

Recent advances by Chinese AI vendors, including Alibaba, Moonshot AI and Z.ai, have highlighted the growing capabilities of open models. Nicholson said frontier-model companies have recognized that open models are approaching the quality of their systems while remaining available at substantially lower cost.

“They are under siege by a lot of the emergence of all these open weight models,” said Lian Jye Su, an analyst at Omdia, a division of Informa TechTarget. “It is a business risk that they are facing at the verge of the IPO.”

Su said Chinese vendors have also argued that calls for regulation are particularly associated with the United States because China is already introducing rules for AI. China recently introduced rules intended to prevent minors from forming relationships with AI chatbots.

“Maybe it’s for the IPO, maybe it’s in [Anthropic’s] interest to slow things down, but it’s still a very important thing because the U.S. fundamentally doesn’t have that AI governance regulation as compared to China,” Su said.

He added that Anthropic also views open source as a potential risk because open source is a decentralized and highly public technology framework with less governance.

Hedging against regulatory uncertainty

For enterprises, uncertainty over whether future regulations will apply to open-source AI, combined with geopolitical tensions involving Chinese models, could lead some companies to continue using frontier models despite their higher cost. Nicholson said businesses may do so to hedge against the possibility that regulation will affect other options in the future.

“[Enterprises are] willing to pay for certainty,” he said.

Kompella said enterprises should respond to the regulatory uncertainty with a proactive approach. “Governance has to become an operating capability, not a checkbox exercise,” he said.

He said companies cannot treat AI governance as a policy document that is reviewed only periodically.

“That is inadequate for systems that can access data, write code, use tools, make decisions and act autonomously,” Kompella said.

He advised enterprises to prioritize early-access and preview programs to stay closer to frontier capabilities, particularly in AI cybersecurity.

“Cyber preparedness becomes central,” Kompella said. “As agentic systems gain stronger offensive and defensive capabilities, security teams need to assume that AI will compress both attack and response cycles.”

Source: AI Business