Anthropic and Nvidia Leaders Clash Over Artificial Intelligence Safety Pace at Dreamforce

Anthropic Chief Executive Dario Amodei and Nvidia Chief Executive Jensen Huang offered conflicting visions for artificial intelligence governance and model deployment speed during Salesforce's annual Dreamforce conference in San Francisco, as first reported by CNBC Business.

Amodei addressed an audience of roughly 12,000 people at the Moscone Center alongside Salesforce CEO Marc Benioff, elaborating on a proposal he published over the weekend that urged model builders to moderate their development velocity. His essay followed warnings from research groups regarding severe systemic risks linked to rapidly advancing systems. Amodei outlined a three-step framework designed to curb the pace of capability growth without undercutting commercial benefits or American technological leadership.

"I think that's the way to lead the industry forward, to set an example, to say that everyone can always be better," Amodei told Benioff during the keynote presentation. His call for restraint received public backing from several prominent industry leaders, including OpenAI CEO Sam Altman, SpaceX CEO Elon Musk, and Google DeepMind Chair Demis Hassabis.

Huang took the stage shortly after Amodei and rejected the notion that technical progress must slow to ensure safety, asserting that market forces already govern responsible behavior without requiring new legal frameworks. "It's a false choice," Huang told Benioff regarding the tension between rapid innovation and risk management. "You could definitely have both at the same time."

Huang placed responsibility directly on individual model providers to judge the readiness of their tools before releasing them into the market. "Run as fast as you can," Huang said during his appearance. "But if you feel at any given point in time the company's out of control or the product's not going to be safe, take a pause and make sure you get it right."

The Nvidia chief executive's Dreamforce appearance followed a session a day earlier at the All-In Summit in Los Angeles, where Huang took a phone call on speaker from President Donald Trump. Trump voiced concern over critics trying to block artificial intelligence data center infrastructure projects, to which Huang replied, "We're not going to let that happen, sir."

Addressing reporters on Tuesday, Benioff offered a pragmatic view on the debate among his key hardware and software partners. "If they feel like they should slow down, then they should slow down. If they feel they should speed up, they should they should speed up. And then they should be held accountable," Benioff said. The conference, which draws around 50,000 people to downtown San Francisco annually, also featured product announcements, including a new reasoning model for Salesforce's Agentforce platform built on Nvidia's Nemotron open-weight models, as well as an expanded partnership between Salesforce and Anthropic around its Claude assistant.