At the Salesforce Dreamforce conference in San Francisco on September 15, 2026, Nvidia CEO Jensen Huang stated that AI safety is an “engineering problem” and that market forces and existing laws are sufficient for oversight.
Huang asserted that developers should be trusted to manage safety protocols as they would for any other industrial or digital product. He suggested that existing laws and market forces provide sufficient oversight.
Addressing the ‘Alien Mind’
Huang’s remarks directly addressed a September 6, 2026, essay titled An Alien Mind by OpenAI chief scientist Jakub Pachocki. In that piece, Pachocki suggested that as models gain complex “chain of thought” reasoning, they become less like predictable tools and more like foreign intelligences that could eventually become difficult for humans to monitor or manipulate.
Huang addressed the “alien mind” concept by framing the discussion around technical safety. He cautioned G20 representatives earlier this month against regulating based on “theoretical harms,” a sentiment he maintained at Dreamforce by asserting that existing laws and market forces provide sufficient oversight to ensure products remain beneficial.
A Growing Rift in Silicon Valley
The stance taken by Nvidia, which currently holds a market capitalization of approximately $5.1 trillion, highlights an ideological and economic schism within the tech sector. While hardware providers like Nvidia benefit from a high-speed development cycle, software-focused competitors have begun calling for the brakes.
Speaking at the same conference, Anthropic CEO Dario Amodei offered a starkly different perspective. Amodei advocated for a slowdown in development and the introduction of third-party safety assessors to verify the safety of advanced models before they reach the public. This “slow-and-steady” approach contrasts sharply with Huang’s view that the single worst outcome for any country is being left behind in the AI race.
Political and Economic Alignment
Huang’s pushback against regulation aligns with broader political movements aimed at maintaining American dominance in the sector. On September 14, 2026—just one day before the Dreamforce appearance—President Donald Trump reportedly called Huang during the All-in Summit to offer support for this deregulatory stance, reportedly labeling certain AI safety concerns as a “hoax” designed to stifle innovation.
As the debate continues, the industry remains split: hardware giants view AI as a tool to be perfected through engineering, while software labs increasingly warn of a technology that may eventually outpace human oversight. For Huang, the solution is simple: allow the engineers who built the technology to be the ones who secure it.
