Core Summary

Nvidia CEO Jensen Huang publicly stated that the AI industry “does not need new laws” and emphasized that existing legal frameworks are already sufficient to address the risks AI poses. His position contrasts sharply with recent statements from OpenAI CEO Sam Altman and Anthropic calling for stronger AI safety measures, highlighting a fundamental divide within the AI industry over the right regulatory path.

Event Details

Huang’s Core Argument

According to the BBC, Huang said in a recent public interview that the AI industry has already established self-regulatory mechanisms across ethics review, content filtering, and data privacy, and that introducing new dedicated laws could “stifle innovation” and leave regulators unable to keep pace with rapid technological evolution.

He further pointed out that existing antitrust, consumer protection, data privacy, and copyright laws already cover most risks AI technology brings. In his view, regulators should “let current laws catch up to the AI era” rather than “reinvent laws for AI.”

The Industry’s Internal Divergence

OpenAI’s Position: OpenAI CFO Sarah Friar told CNBC she is “not worried about slowing AI development,” while OpenAI has publicly backed bipartisan congressional drafts on AI safety assessments.

Altman’s Two Demands: Sam Altman has publicly identified two priority regulatory goals: third-party safety assessments for frontier AI models and dedicated early-warning systems for AI-generated biothreat risks.

Anthropic’s Position: CEO Dario Amodei has long championed a “responsible scaling policy,” calling for government-level licensing of training and deployment of ultra-large models.

Huang’s Counterargument: He argues AI companies already have strong commercial incentives to ensure AI safety, since any major incident would destroy the company’s commercial value, making government intervention unnecessary.

The Complex Congressional and White House Battle

Politico reported that Congress is engaged in fierce debate over an AI regulatory framework:

  • House Speaker Johnson has played down AI existential-risk warnings, declaring “you’re not all going to be dead in 10 years,” and opposes using “existential risk” as a basis for an AI pause.
  • Al Jazeera reported that the divide between the “pause camp” and the “acceleration camp” in Congress is deepening; the former fears AI runaway will threaten human survival, while the latter worries that over-regulation will cost the US its edge against China.

Panoramic Analysis

First, the AI industry’s internal “regulatory philosophy war” has gone public. Huang represents the “industry self-regulation + existing laws” approach, essentially the free-market camp; OpenAI and Anthropic represent the “frontier models need dedicated regulation” camp, essentially the precautionary-principle camp. The substance of the debate is not whether to regulate, but who regulates, who is regulated, and how far. Behind the three positions lie differences in business models and incentive structures: as a compute supplier, Huang naturally wants regulation minimized to preserve market demand; while players built around model capabilities want rules that benefit them.

Second, the “compute-as-power” logic is reshaping AI regulation. Nvidia, as the world’s most important AI compute supplier, has structural influence on global AI development. If Huang’s “no new laws” position prevails, the focus of AI regulation will remain on the “application layer” rather than the “compute layer,” significantly undermining the legitimacy basis for national compute export controls (such as the US chip ban on China).

Third, competition with China is becoming the “veto power” of AI regulation. Any US congressional proposal to strengthen AI regulation will inevitably face the challenge of “whether it will set the US back in the China race.” This structural imbalance means US AI regulation will likely remain “low density” for the foreseeable future.

Fourth, global AI governance will move toward “multipolar fragmentation.” The US may maintain low-density regulation, the EU will form strict regulation through the AI Act, and China will build a national-security-centric regulatory system. The three sets of rules will create “regulatory trade barriers,” raising compliance costs for global AI companies and producing a new geopolitical technology landscape.

Multiple Perspectives

Nvidia’s Position: Huang emphasizes that “AI is a general-purpose technology like electricity or the internet, and over-regulation will kill its social value.” He advocates building codes of conduct through industry self-regulatory bodies (such as the Partnership on AI) rather than through legislation.

OpenAI’s Position: Altman has repeatedly called for “frontier AI models to undergo independent third-party audits” and has endorsed the Future of AI Act being pushed in Congress. He acknowledges “AI does carry risks” but opposes the extreme position of “trading safety for a pause.”

Anthropic’s Position: Dario Amodei has long called for an “AI safety tier system,” proposing mandatory safety assessments for frontier models whose training costs exceed a specified threshold, similar to the safety regulatory model used in the nuclear industry.

Academic View: Most scholars at Stanford’s AI Research Institute and MIT’s Future Technology Institute favor “moderate regulation”—tiered regulation at the application layer while keeping fundamental research open.

Government Position: The White House Office of Science and Technology Policy has issued the “AI Bill of Rights Blueprint,” but it lacks legislative force. The EU AI Act has entered the implementation phase and will impose real constraints on global AI companies.

Editor: GoodInfo Global News Team