Nvidia (NVDA) CEO Jensen Huang Dismisses AI Cybersecurity Scare as Manufactured Demand

Huang Tells Goldman Sachs Audience Cyber Fear Is Manufactured

Nvidia (NVDA) chief executive Jensen Huang told a San Francisco investor audience on Thursday that the wave of alarm surrounding AI-driven cybersecurity threats is, in his assessment, engineered demand — and that the companies selling the cure are amplifying the scare. Speaking at the Goldman Sachs Communacopia + Technology Conference, Huang argued that security vendors talk loudly about danger because product launches are approaching, and his remarks drew a room of institutional investors tracking how AI budgets form for next year. “The reason why there’s so much conversation today about cybersecurity is because the industry is getting ready to launch some products, and what better way to create demand than to create a problem?” he told the audience. He pressed the point with a rhetorical question of his own, asking who would not want a hysterical market queuing around the corner for its product. Notably, Huang did not dismiss the sector itself: he called cybersecurity the next major use case for artificial intelligence and a strong business opportunity, making clear his criticism targets the fear campaign rather than the market. The remarks landed days after Huang declared that human-level AI has already arrived, a call he made before OpenAI advanced any comparable claim of its own — a sequence we examined in our earlier coverage of his human-level AGI statement. The timing is awkward for crypto audiences. Security vendors and AI laboratories have warned for months that AI-enabled attacks are accelerating, and crypto platforms have absorbed much of that damage, which is why Huang’s framing reads to many in the space as textbook FUD management from the industry’s dominant compute supplier. NVIDIA (NVDA) sits at the center of the AI buildout, giving its chief executive’s characterization of threat narratives outsized influence over how investors read the security spending cycle.

Crypto Attack Data Complicates the Pitch

The incident record from the same period tells a harder story than the sales-pitch framing suggests. Blockchain security firm SlowMist’s incident data shows attacks on crypto platforms climbed roughly 50% in the first half of 2026, even as the total value of stolen funds shrank — more intrusions yielding smaller hauls, a pattern consistent with automated tooling doing the initial probing. The same dataset points to faster exploit cycles against DeFi protocols, where machine-speed discovery shortens the window between a vulnerability surfacing and funds being drained. The corporate alarm predates Huang’s comments by weeks: more than 100 companies signed an open letter in August warning that AI-enabled attacks would spread within months, with OpenAI and Anthropic both adding their names to the document. Budgets followed the warning. CrowdStrike stock jumped 20.5% in late August after record earnings showed AI threats lifting enterprise security spending — evidence, for now, that the fear campaign Huang describes is converting into real orders. Security researchers see the same pressure inside crypto. OpenZeppelin co-founder Manuel Aráoz argues that AI agents probe smart contracts faster than human reviewers can clear them, a pace problem that strains every layer of the stack from DApp code audits to private key custody routines. Exploit teams no longer need to sleep, and the asymmetry between machine-speed attack and human-speed review widens each quarter. None of this disproves Huang’s underlying point that vendors monetize anxiety; the CrowdStrike rally demonstrates exactly that dynamic at work. But it does mean the threats themselves are not theater — the platforms absorbing them are processing a rising number of live incidents, and upcoming loss reports will show whether frequency without severity still moves the market. Readers tracking the market in real time can follow live spot and futures prices on Gate.

COINOTAG’s read is that both readings hold at once. Vendors do package fear into product launches, yet the attacks keep landing, and the most load-bearing document in this dispute is the August open letter itself — a signed, first-party warning from more than 100 companies, including the labs building the models, that AI-enabled attacks would spread within months. That sits awkwardly beside Huang’s demand-creation argument. For Nvidia the stakes are commercial: fresh off its $12.93 billion Hugging Face acquisition, the company sells compute to both sides of the AI security race, while analyst Tom Lee treats NVDA’s earnings-day jump as a market health signal and rival Enflame’s 234% Shanghai debut shows the competitive clock ticking. Crypto security teams will judge the rhetoric by next quarter’s loss reports.