Google Brain Co-Founder Andrew Ng Dismisses AI Extinction Risks

Google Brain co-founder Andrew Ng has dismissed recent warnings that artificial intelligence poses an existential threat to humanity as "much more science fiction than science," arguing that the industry is leveraging catastrophic scenarios to influence regulation and secure competitive advantages [1]. The pushback arrives as a growing divide emerges between researchers calling for slower development and industry leaders who maintain that current safety concerns are being manufactured to protect established market positions [1, 3].

The battle over regulation and moats

Ng, who previously helped launch Google Brain and Coursera, stated he was "quite dismayed" by the recent surge in warnings from AI researchers and executives [1, 3]. He contends that the industry is repeating a pattern seen in 2023, where large companies hype existential fears to "pull up the ladder" against new market entrants through restrictive policy [2]. Investor Steve Eisman supported this view, suggesting that because major AI firms lack durable "moats"—or sustainable competitive advantages—they are manufacturing a crisis to justify regulatory barriers that would insulate their businesses from competition [1, 3].

This skepticism follows a high-profile resignation from an Anthropic researcher on September 8, which sparked a wave of public concern regarding the pace of super-intelligent AI development [2]. While executives like OpenAI’s Sam Altman and Anthropic’s Dario Amodei have advocated for greater safeguards, others including Nvidia CEO Jensen Huang and Meta CEO Mark Zuckerberg have publicly opposed calls to slow the industry's progress [1].

Practical engineering vs. existential fear

Despite the heated rhetoric, the European Union has already codified both sides of the debate into law. The EU AI Act’s list of systemic risks includes both concrete engineering challenges—such as cybersecurity and chemical or biological threats—and abstract "loss of control" scenarios, such as models making copies of themselves or failing to align with human intent [2].

For now, the regulatory environment is shifting toward mandatory compliance. Tech chief Henna Virkkunen confirmed this month that EU law now requires companies like OpenAI and Anthropic to assess the risk of losing control of their models, effectively forcing firms to treat these "science fiction" scenarios as practical engineering requirements [2].

What to watch

• Regulatory Enforcement: Monitor how the European Commission applies its "loss of control" assessment requirements to frontier AI models in the coming months [2].
• Industry Lobbying: Watch for further public statements from AI labs as they navigate the tension between advocating for safety and maintaining their competitive standing against new entrants [1, 2].
• Public Sentiment: Observe whether the trend of tech workers mocking extinction warnings via social media memes continues to influence the broader public discourse on AI safety [2].

The core disagreement remains whether the industry is genuinely addressing the risks of super-intelligence or simply using the specter of catastrophe to consolidate power. With European law now mandating the assessment of existential risks, the debate has moved from theoretical discussion to a compulsory operational burden for major AI developers [2].