Breaking

Nvidia CEO Jensen Huang rejects AI-doom warnings as slowdown debate widens

September 14, 2026 at 20:13 EDT

Foundation Models
Policy & Regulation
Infra & Chips

Nvidia CEO Jensen Huang used a Sept. 14 appearance at the All-In Summit in Los Angeles to reject catastrophic AI predictions as unsupported by science, putting one of AI's most important hardware suppliers publicly against calls for a frontier-model slowdown. President Donald Trump joined the interview by phone and amplified the same message, turning a technical safety dispute into an overt policy fight.

Sept. 14, 2026 / Nvidia

Huang rejects AI doom warnings

The Nvidia CEO sided against calls to slow frontier AI, while the safety camp pushed for outside evaluators and coordinated pacing.

The debate moved in three beats

Slowdown call

A three-step pacing plan put frontier safety back at the center.

→

Live pushback

Huang rejected catastrophe forecasts as unsupported by science.

→

Policy fight

The argument now spans safety, chips, data centers and U.S. competitiveness.

Public resistance is already concrete

Opposition to local AI data centers is higher than opposition to nearby nuclear plants.

71%

AI data centers

53%

Nuclear plants

Each block represents about 10 percentage points.

3

steps in the pacing proposal

6-12

months in the cited risk window

36+72

Grace CPUs and Blackwell GPUs in GB200 NVL72

Pro-build reading

Continued compute expansion supports model deployment, industrial adoption and the U.S. lead in AI.

Safety-first reading

Voluntary pacing needs enforceable commitments, or it risks becoming industry self-protection.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.