NVIDIA CEO Jensen Huang has cautioned technology leaders to carefully balance discussions about AI risks, warning against turning legitimate concerns into unfounded panic. On March 19, when asked how AI startup Anthropic could better handle its contract dispute with the U.S. Department of Defense, Huang stated:

It is important to remind people of the potential risks of this technology. Reminders are beneficial, but intimidation is not, because this technology is too important for us.

In a panel discussion later released as a podcast, Huang expressed support for Anthropic's mission and safety principles but noted that the industry sometimes exaggerates AI risks. He emphasized that the greatest AI risk to U.S. national security is not the technology itself, but the potential slowdown in AI adoption due to public anger, fear, or paranoia.

Huang explained:

AI is not a living organism, an alien species, or conscious. It is computer software. Making extreme, catastrophic claims without evidence can cause more harm than people realize.

At the same time, Huang expressed optimism about Anthropic's commercial prospects, predicting the company's annual revenue could surpass $1 trillion by 2030. He suggested that CEO Dario Amodei's own forecasts tend to be conservative.

Despite Anthropic's ongoing dispute with the government, Huang remains bullish on its business potential. During the discussion, he indicated his belief that Anthropic's annual revenue could exceed $1 trillion by 2030, adding that Amodei has historically been conservative in his projections. This demonstrates Huang's strong confidence in Anthropic's core business value, contrasting with current political controversies surrounding the company.

Anthropic is currently challenging the U.S. Defense Department's designation of it as a "national security supply chain risk" through legal channels. The conflict stems from Amodei's insistence that contracts explicitly prohibit using its products for domestic surveillance of U.S. citizens and developing fully autonomous weapons systems.

The previous administration responded strongly to these requirements, labeling Anthropic a supply chain risk and moving to exclude it from government programs. Anthropic has filed a lawsuit regarding this matter, and the standoff continues. Anthropic is an important customer of NVIDIA and the developer of the Claude series of chatbots.