This article is part of the weekly Technology newsletter, which is sent every Friday. If you want to sign up to receive it, you can do so at this link.
“This guy is a monster,” said Donald Trump in front of Sundar Pichai, CEO of Google. "And no one knows him. What a great life! To be a monster and not have to go through all this," Trump continued pointing out to the journalists who follow him everywhere. Meanwhile, at his side, Pichai waited to speak with an incredible face of resignation. Before finishing, Trump joked about his Indian name: “Sundarrrrr is his name.” When it was Pichai's turn to speak, he hid it as best he could: "Hopefully, it will be a better letter of introduction than 'I'm a monster,'" he said.
The scene occurred after Trump's meeting this week with the big leaders of the artificial intelligence (AI) industry. It was at the time of statements to the press in the gardens of the White House. The meeting had served to sign an agreement that said that large laboratories must self-regulate and forget that the Government is going to intervene to stop the progress of the new models.
It was basically a group of extremely rich people behaving like children who were afraid of a bully. There are several videos showing nervous laughter, awkward looks, and absurd jokes of Trump pleasing some of the richest people in the world.
The most viral fragment on networks is a cut of about 70 seconds where the nerves of Dario Amodei, co-founder of Anthropic and the biggest proponent that AI should slow down, are clearly visible. Trump doesn't agree with him, but they seem to get along well now. Before Amodei spoke, Trump told reporters: “Anything he says [pointing to Anthropic] is fine.” Then he grabbed him by the arm and whispered, “Be careful.”
Amodei is clearly a peculiar gentleman. But his excessive laughter literally lasted a second and a half. It's epic, pure theater. Then Amodei spoke for a few seconds, without further ado. Meanwhile, Trump approached Mark Zuckerberg, from Meta, as if he were the mischievous one hiding. He treated them like children, he played jokes in their ears. Musk's face in the background was clearly spiteful.
Trump then spoke again and mentioned the CNN network, which maintains a pulse with the president these days, and said: “CNN, fake news.” It was a poker face moment for the executives. Trump was messing with a huge media and they were there to the side partially condoning the affront. There was no need to smile and the best thing was to disappear.
It seemed to Zuckerberg that everything didn't matter to him, that he was so above good and evil that he couldn't even avoid a smile due to the discomfort. It is seen more in this other image, where Amodei was fleeing:
Before these ridiculous scenes, there was someone at that AI boss meeting with Trump who usually goes unnoticed, but who was to the president's right. It was Jensen Huang, CEO of Nvidia, the most valuable company in the world. The text of the agreement says what Huang thinks: that AI companies cry less and solve their problems alone. Zuckerberg thinks that too.
A few days ago, Huang was on a podcast for almost 2 hours talking about AI. He told why there is no need to be afraid. Huang co-founded Nvidia in 1993, which designs chips used for AI. He is the seventh richest person in the world, just behind Mark Zuckerberg. He has never been very talkative and always has a rather serious pose.
Huang believes that Amodei, Altman and others are exaggerating and should stop scaring. Their argument is very simple: if they really suspect that their product is dangerous, don't launch it. What other company would put something on the market that can cause death and destruction?
He seems very confident and calm when speaking. He is also very vehement about what AI is ("it's just software") and that it will not bring more problems than other historical revolutions: "They should not launch the product. It's that easy. If you are going to make an autonomous car and there is a complicated situation, and as engineers we have no idea how to solve it, because these cars are not programmed, they are trained, what is the answer? Don't bring it up. It's an engineering problem. You have to know how to solve it. Now, if they also say that there is no way to contain the experiments, that "When we test the AI models, they will escape and harm the world, so the answer is to close the laboratories," explains Huang, in a very non-apocalyptic tone, far from other large companies, but very close to the White House.
