GOP Senators Urge Intelligence Officials to Boost Assessments of China’s AI

Two Republican U.S. senators have urged senior intelligence officials to strengthen assessments of China’s artificial intelligence (AI) capabilities.

Sens. Jim Banks of Indiana and Tom Cotton of Arkansas sent a letter on Thursday to senior officials in the intelligence community, including Director of National Intelligence Tulsi Gabbard, who plans to leave office on June 30 to support her husband, who has been diagnosed with a rare form of bone cancer.

The full text of the letter, which was first shared with The Hill, was not available on official congressional websites as of publication.

The Epoch Times reached out to Banks’ office for comment and did not receive a response before publication time.

The senators have a record of addressing technological competition with China. On 30 April, Banks introduced the AI Overwatch Act with co-sponsors including Cotton.

The bill aims to prevent advanced chips from fueling China’s military. It was referred to the Senate Committee on Banking, Housing, and Urban Affairs.

“We are in an AI arms race with the Chinese Communist Party that’s powered by advanced chips and the winner will determine the future of global military superiority,” Banks said in a statement on the bill.

“Our most advanced AI technologies must be kept out of the hands of adversaries like Communist China,” Cotton added.

U.S. policy has restricted exports of advanced semiconductors to China since 2018 to limit military applications of AI.

The Bureau of Industry and Security within the Department of Commerce has updated these controls, including a global AI diffusion rule in January. Official records describe the goal as hampering China’s military modernization while not seeking broader economic decoupling.

On Feb. 26, Banks and Cotton joined in signing a letter urging stronger outbound investment restrictions in AI and semiconductors, highlighting risks of U.S. capital supporting China’s defense sectors.

China has continued developing its AI sector despite restrictions.

U.S. government reports note Chinese efforts to acquire chips through smuggling and third countries. Congressional oversight has repeatedly addressed these issues in hearings and legislation.

The senators’ latest letter follows the April limited release of Anthropic’s Claude Mythos Preview model. Anthropic’s official system card states that the model shows advanced capabilities in identifying security vulnerabilities. Access was restricted to a handful of companies and government agencies due to potential risks.