Nvidia CEO declares 'AGI has arrived,' declares '0% chance' it ends the world, 13 days apart.

As it ran on the front

On September 7th, Nvidia's chief executive Jensen Huang posted that "AGI has arrived" — artificial general intelligence, the machine-capability threshold researchers have spent decades warning could reorder or end human civilization — arriving, in his telling, inside a chatbot update from OpenAI. Seven days later, in Los Angeles, President Trump called Huang's phone while Huang was onstage and told him, in front of the room, that talk of slowing or regulating that technology was excessive. The same week Trump told reporters the AI-safety warnings themselves were a "hoax," a word plain enough that House Speaker Mike Johnson went on camera to share his own thoughts on it. The next day, September 15th, Trump's AI adviser David Sacks sat for CBS and dismissed both the warnings and the calls for more regulation attached to them. But Sacks did not say there was nothing to worry about. He said, in the same breath, that it is "on them" — the companies — to make their own products safe. Not a federal agency. Them. The company gets to grade its own homework.

Which brings us back to Huang, on the same network, with the same correspondent, five days later: "There is 0% chance" AI ends the world by 2030, he told CBS's Jo Ling Kent on September 20th — the same executive who, thirteen days earlier, had announced the arrival of the very intelligence threshold the doomsday conversation is about. The warnings, Huang said, are "quite frankly too much drama, irresponsible" and "not grounded in facts." The filing, in order: AGI arrived on the 7th. Zero risk was certified on the 20th. Nothing changed in between except who was talking.…