Nvidia’s chief calls AI safety an engineering problem as the industry argues over the rules
Jensen Huang rejected the idea that security failures are an unavoidable cost of the technology, while the president called warnings about AI a hoax.
In an interview with CNN’s Anderson Cooper, the Nvidia chief executive, Jensen Huang, argued that AI safety is fundamentally an engineering challenge, The Hill reported1.
He also rejected the argument, associated with the OpenAI chief executive, Sam Altman, that cybersecurity breaches may be an unavoidable consequence of artificial intelligence, Storyboard18 reported2. Separately, Huang predicted that artificial intelligence will become an essential part of every business and that no company will be exempt3.
The positions are not reconcilable on their face. One treats harm as a design defect to be engineered out, the other as a residual cost of deployment1,2.
Huang’s claim is about method rather than outcome: he is arguing that the failures critics point to are tractable with better engineering, not that they are unimportant1. Storyboard18 reported the exchange as a direct rejection of the trade-off framing rather than a difference of emphasis2.
Get AiM Weekly, free.
The market backdrop
The remarks landed in a week of selling. Wall Street sees the AI trade as intact despite the latest rout, with one characterisation of the episode as just another scare, according to reports carried by AOL and Yahoo Finance4,5. Huang sought to ease concerns by expressing optimism about CoreWeave, and the reports note that Nvidia’s backing of the company is significant4,5.
Tech Insider reported that President Donald Trump called warnings about AI a hoax, and recorded a 0.9% fall in the Nasdaq, while listing Altman under a safety-first stance6. That is an aggregator’s summary rather than a transcript, and it should be read as such.
Huang’s posture has been consistent. Yahoo Finance noted that his September post on X declaring that AGI has arrived raised eyebrows across the markets7.
Why it matters to investors
The safety argument is a proxy for a regulatory one. If harm is an engineering problem, the remedy is product design and the cost sits inside the companies’ own development budgets1. If it is an unavoidable byproduct, the remedy is rules, and the cost sits on top of capital spending already under scrutiny for how it is financed2.
The White House position matters to the timetable rather than the substance: a president describing the warnings as a hoax is unlikely to accelerate federal rulemaking6. Watch whether the chip and model companies continue to argue in public, because a split among them is what usually precedes legislation written by someone else1,2,7.
The nearer test is commercial. If Nvidia’s optimism about CoreWeave is read as support rather than opinion, it blurs the line between a supplier’s outlook and a customer’s balance sheet, and that is a judgement investors have to make for themselves4,5.
Sources
- New research warns of AI’s impact on children’s mental health - The Hill, thehill.com (2026-10-18)
- Nvidia CEO Jensen Huang rejects Sam Altman’s AI cybersecurity trade-off argument, storyboard18.com (2026-10-18)
- Nvidia CEO Jensen Huang says every company will eventually become an AI ... - Business Upturn, businessupturn.com (2026-10-18)
- ‘Just another scare’: Wall Street sees the AI trade strong despite tech’s latest rout - AOL, aol.com (2026-10-18)
- ‘Just another scare’: Wall Street sees the AI trade strong despite tech’s latest rout, finance.yahoo.com (2026-10-18)
- Trump Calls AI Warnings ‘HOAX’, Nasdaq Drops 0.9% [2026] - Tech Insider, tech-insider.org (2026-10-18)
- Breakfast News: The Sunday Edition - Yahoo Finance, finance.yahoo.com (2026-10-18)
Assembled by Edwin, my AI assistant powered by Claude, from the public excerpts of the outlets numbered above. No human wrote or checked it before publication, so read the sources before you act on it.