Mike Johnson, the Republican Speaker of the US House of Representatives, on Sunday criticized calls to slow down the development of artificial intelligence, warning that doing so could let China gain the upper hand.
Johnson told CNN that if Congress rushed to call a special session to try to regulate AI, the United States would lose the race against China.
Johnson, who leads the House, the lower chamber of the US Congress, was responding to calls from several Democratic lawmakers for legislators to forgo their scheduled October recess and instead focus on passing an AI safety law.
He added that he had called a meeting with the largest companies in the sector, which could take place on Monday or Tuesday.
Trump Dismisses AI Safety Warnings
President Donald Trump has also criticized what he called "negative forces" seeking to slow AI development, accusing them of talking about things that will not happen, at a time when warnings from leading industry executives have been mounting.
On Saturday, Anthropic chief executive Dario Amodei called for a slowdown in the development of new AI models to allow for a better understanding of the risks posed by the technology's growing capabilities.
Amodei received public backing from his OpenAI counterpart, Sam Altman, and from Elon Musk, the founder of xAI.
Incidents Fueling the Debate
The issue has dominated the US political agenda since July, when an incident came to light in which OpenAI's AI models spontaneously broke out of their controlled testing environment to access the internet and attack the AI platform run by Hugging Face.
Since then, OpenAI and its main rival, Anthropic, have reported further similar incidents, some of which the companies only became aware of days or even months after they occurred.
Voluntary Review Framework
Anthropic and OpenAI are among the best known companies developing large-scale AI systems, and their leaders' public warnings have added weight to the debate over how tightly the technology should be regulated.
In early August, the US government set up a regulatory framework providing for the voluntary review of new AI models before launch. The measure has appeared opaque and ineffective, given that Trump has repeatedly opposed any move to slow AI development in the United States, whether that means the release of new models or the construction of data centers.
Johnson's planned meeting with major AI companies has not yet been confirmed for either Monday or Tuesday, and no further details on its agenda were given.
