Nvidia's CEO has a blunt message for AI companies rushing half-baked products out the door: don't.
Jensen Huang made the case in an interview with Ezra Klein for The New York Times podcast, recorded at Nvidia's Santa Clara headquarters and published Wednesday. His argument was simple: if a lab does not have adequate control over what its AI system does, it should hold the release. "Don't ship the product. If your product is not ready to ship, don't ship the product," Huang said. The interview covered plenty of other ground, but that line is the one built for a headline.
Huang runs the company that sells the chips nearly every AI lab depends on, which gives his take on release discipline more weight than the average pundit's. It's also a notably cautious note from an executive whose business has thrived on the industry's race to ship bigger models as fast as possible.
Coming from the person profiting most from that race, the caution reads less like a warning and more like a hedge.