On March 20, Bloomberg reported that Nvidia CEO Jensen Huang said on Thursday that technology leaders need to act cautiously to avoid triggering public panic about AI. Jen-Hsun Huang made the remarks in response to Anthropic’s dispute with the U.S. military over the use of AI. During a panel discussion at the Nvidia Technology Conference on Thursday, he was asked how Anthropic could have better handled thorny contract negotiations with the Pentagon.

Jen-Hsun Huang

"The ability to draw attention to this technology is very good," Huang said. "Warning is good, but creating panic is not so good because this technology is so important to us."

Behind Huang Renxun's suggestion is that he believes that the biggest national security risk facing the United States in the field of AI is that Americans who are too angry, fearful, or paranoid will cause the country to adopt the technology slower than its competitors. As an important customer of NVIDIA and the developer of Claude chatbot,Anthropic is currently embroiled in a conflict with the Trump administration over its desire to impose restrictions on the military use of its AI tools..

Anthropic's relationship with the Pentagon broke down last month after Anthropic CEO Dario Amodei insisted on adding a clause in the contract banning its products from being used for domestic surveillance of U.S. citizens and for fully autonomous weapons systems. In response, the Trump administration declared Anthropic a supply chain risk and moved to terminate all of its collaborations across the U.S. government, measures the company is currently challenging in court.

Despite Anthropic's dispute with the U.S. military, Huang remains optimistic about Anthropic's financial future. During the panel discussion, Huang said he believed Anthropic could surpass $1 trillion in revenue by 2030. He also added that he believed Amodei's forecast was too conservative.

When asked specifically how he would handle a dispute with the military if he were on Anthropic's board, Huang first expressed his appreciation for the company, including its focus on safety and security, but then said the industry must be careful to avoid stoking unnecessary public fear about AI tools.

"It is not a living organism. It is not an alien object. It has no consciousness. It is just computer software," Huang said. "To make some quite extreme, quite disastrous, and without evidence statements, the harm may be far greater than people imagine."

When asked about the strategic risks of concentrating advanced chip production in Taiwan, Huang said that the AI ​​manufacturing supply chain should be diversified and mentioned the deployment of production in South Korea and Japan as well as the United States.

As of press time, an Anthropic spokesperson had not commented.