The AI Debate at Davos
Fault lines emerge among the global elite.
The World Economic Forum wrapped up Friday in Davos, and if you followed the coverage, you might think the whole thing was about Greenland. When President Trump wasn’t dominating the news cycle, however, AI was the through-line connecting nearly every conversation, from labor markets to national security to the nature of human intelligence itself.
What made this year’s summit different wasn’t the hype, of which we’ve had plenty, but the sharpening disagreements among people who are actually building, regulating, and warning about these systems. The week produced a remarkable series of clashes: tech optimists versus labor economists, chipmakers versus AI safety advocates, philosophers versus billionaires.
The AGI Predictions Get Bolder
Elon Musk made his Davos debut on Wednesday, appearing alongside BlackRock CEO Larry Fink with a characteristically bold opening gambit: we might have AI smarter than any individual human by the end of this year, or no later than next year. By 2030 or 2031, he suggested, AI could surpass the combined intelligence of all humanity.
Musk has been making versions of this prediction for years, but the specificity is new, and the timeline keeps compressing. He also announced that Tesla’s Optimus robots are already performing basic operations in factories and will tackle complex tasks by year’s end, with consumer sales planned for 2027.
The constraint now, Musk argued, isn’t compute but electricity: “Very soon, we’ll be producing more chips than we can power.” He pointed to China’s massive solar buildout as a sign of the infrastructure race to come.
“Like Selling Nuclear Weapons to North Korea”
The week’s most striking clash came from an unlikely source: Anthropic CEO Dario Amodei publicly criticizing Nvidia, one of his own company’s major investors, over the Trump administration’s decision to allow H200 chip exports to China.
“I think it would be a big mistake to ship these chips,” Amodei said in an interview with Bloomberg. “I think this is crazy. It’s a bit like selling nuclear weapons to North Korea.”
The comment landed hard, given that Nvidia has invested billions in Anthropic. But Amodei’s argument was blunt: the U.S. leads China in chip manufacturing by years, and export restrictions are “the thing that is holding them back.” Lifting them, he suggested, carries “incredible national security implications.”
Nvidia CEO Jensen Huang, also at Davos, has long argued the opposite: that China will simply develop its own chips if denied access to American ones. He’s reportedly planning a trip to Beijing later this month to attend company parties ahead of the Lunar New Year.
The rift illustrates a growing divide in the AI industry between those who see geopolitical competition as the defining constraint and those who see it as a distraction from the real work of deployment and scaling.
The Labor “Tsunami”
IMF Managing Director Kristalina Georgieva delivered the week’s starkest warning about AI’s impact on work during a Friday panel on the global economic outlook, describing AI as “a tsunami hitting the labor market.”
The numbers she cited are significant: 60% of jobs in advanced economies will be affected by AI, whether enhanced, eliminated, or transformed, with the global figure at 40%. The distribution is what concerns her most. One in ten jobs in advanced economies has already been enhanced by AI, often resulting in higher pay, which sounds like good news until you consider the other side of the ledger.
The tasks being eliminated are disproportionately entry-level. “Young people searching for jobs find it harder to get to a good placement,” Georgieva said, while middle-class workers whose roles aren’t enhanced by AI are getting “squeezed” as wages grow at the top.
“Wake up,” she told the audience. “AI is for real, and it is transforming our world faster than we are getting a handle on.”
This framing is more grounded than the usual “robots will take all the jobs” discourse. Georgieva isn’t predicting mass unemployment; she’s describing a restructuring of labor markets that could hollow out the middle while benefits concentrate at the top, unless policymakers act.
Worker anxiety is already spiking in response to these dynamics. According to Mercer’s Global Talent Trends report, employee concerns about AI-driven job loss have jumped from 28% in 2024 to 40% in 2026, and a Stanford study found a 16% relative decline in employment for recent graduates in AI-exposed roles, while experienced workers remain stable.
The Counterargument: AI Creates Jobs Too
Not everyone at Davos was sounding alarms. Jensen Huang spent much of his appearances talking about the “largest infrastructure build-out in human history”: the data centers, power plants, and physical infrastructure needed to run AI at scale.
This construction boom, Huang argued, is generating massive demand for skilled trades, with electricians, plumbers, and steelworkers all in short supply and wages rising sharply. “It’s wonderful that the jobs are related to trade craft,” he said, noting salaries have “nearly doubled” in some cases. “Everybody should be able to make a great living; you don’t need a PhD in computer science.”
Microsoft CEO Satya Nadella offered a different kind of optimism, describing an emerging type of knowledge work with new competencies based on how AI reshapes information flows, though he also warned that AI could lose its “social permission” if its benefits remain concentrated among a handful of powerful firms.
Harari’s Warning
Yuval Noah Harari, the historian and WEF favorite, delivered a session titled “An Honest Conversation on AI and Humanity.”
His argument was that AI is not a tool we can control but an agent that learns, adapts, and makes decisions autonomously. Because it operates through language, generating text, code, legal documents, and religious interpretations, it threatens to reshape the very domains humans have used to organize society.
“If laws are made of words, then AI will take over the legal system,” Harari said. “If books are just combinations of words, then AI will take over books. If religion is built from words, then AI will take over religion.”