Anthropic CEO Dario Amodei has once again sounded the alarm on the disruptive potential of artificial intelligence in the workplace. Speaking on BBC Radical with Amol Rajan, he warned that AI could replace all entry-level office jobs in the next one to five years, a timeline much shorter than most expect.

“Specifically, if we look at jobs like entry-level white, you know, I think of people who work at law firms, like first-year associates, there's a lot of document review. It's very repetitive, but every example is different. That's something that AI is quite good at,” Amodei explained.

According to him, AI is no longer a futuristic tool; it is already capable of performing repetitive, variable tasks that form the backbone of many entry-level roles in law, finance, consulting, and administration. Amodei emphasized that many corporate leaders privately view AI not as a way to augment their workforce but as a method to cut costs. “I think, to be honest, a large fraction of them would like to be able to use it to cut costs to employ less people,” he said.

Amodei’s concerns are backed by private discussions he has had with executives who are already integrating AI into operations, many of whom are candid about their intent to reduce headcount. He argues that the technology is improving rapidly, with AI already demonstrating proficiency in tasks traditionally performed by first-year associates, junior analysts, and other entry-level professionals.

This warning is not new. In May, Amodei told Axios that AI could wipe out half of entry-level office roles within five years, potentially raising unemployment to between 10 and 20 percent. Beyond clerical work, he also sees AI taking over software development. “AI could write 90 percent of software code within three to six months and ‘essentially all’ of it within a year,” he said, predicting that human engineers may be limited to defining high-level design parameters while AI handles the detailed execution.

However, opinions among tech leaders are divided. Nvidia CEO Jensen Huang rejected Amodei’s projections, arguing that AI will reshape roles rather than eliminate them. Similarly, Microsoft’s chief product officer, Aparna Chennapragada, underscored the continuing importance of coding skills, stating that computer science remains highly relevant in the AI era. Yet, some leaders, like Ford CEO Jim Farley, share Amodei’s outlook, predicting AI could replace “literally half” of U.S. white-collar workers.

What remains clear is that AI is advancing faster than many businesses, regulators, and employees are prepared for. The debate is no longer whether AI will transform work but how it will do so—whether by displacing jobs en masse or by creating new roles that have yet to emerge. For Amodei, the countdown is already underway. The automation of entry-level positions is not a distant threat but an accelerating reality, and its impact could reshape the workforce in a matter of years.