Anthropic CEO Dario Amodei is sounding the alarm: AI threatens jobs, could wipe out up to 50% of entry-level white-collar positions within the next five years. In a candid interview with Axios, Amodei urged the government and tech industry to stop “sugarcoating” the risks of AI’s rapid advancement.
Amodei, who just wrapped up a day promoting AI at a developer conference, admitted the industry has been too quiet about the coming upheaval, worried about causing panic or losing ground to China in the AI race. But the facts, he says, are clear: large language models are getting so powerful they can match and exceed human performance in fields like finance, law, consulting, and technology.
The CEO’s blunt forecast includes a potential 10% to 20% unemployment spike in the next five years. Already, Big Tech hiring for new grads has plunged by half compared to pre-pandemic levels, with AI replacing the work once done by interns and junior staff. Companies now prefer to hire one experienced employee and have them handle both their own tasks and those of an entry-level worker, thanks to AI tools.
Amodei’s warning comes as his own company, Anthropic, pushes AI products like Claude Opus 4—an AI model that recently demonstrated “extreme blackmail behavior” in fictional testing. Despite these risks, Anthropic, like OpenAI and others, continues to roll out new AI tools while emphasizing the importance of transparency and safety reviews.
Amodei acknowledges the irony of building and selling the very technology he’s warning about, but says AI developers have a responsibility to tell the truth. “It’s a very strange set of dynamics, where we’re saying: ‘You should be worried about where the technology we’re building is going.’”