The head of one of the world’s leading artificial intelligence companies says the technology could eliminate up to half of all entry-level white-collar jobs in the United States within five years, raising the risk of double-digit unemployment unless governments and the private sector begin preparing now.
Dario Amodei, CEO of Anthropic, said AI’s growing capabilities could rapidly displace workers in industries such as finance, law, consulting and technology. In an interview with Axios from his office in San Francisco, he said companies and policymakers are underestimating the pace and scale of the changes underway.
“We could see unemployment shoot up to 10% or even 20%,” Amodei said. “We’re heading toward mass elimination of roles, especially at the entry level. Most people are unaware that this is about to happen. It sounds crazy, and people just don’t believe it.”
Amodei’s warning comes as developers race to improve the performance of large language models (LLM) that can already match or exceed human output in a growing number of tasks. Anthropic, which released its Claude 4 model last week, is among the companies at the center of the AI arms race, along with OpenAI, Google’s Gemini, Microsoft’s Copilot, and Meta.
The rapid expansion of AI capabilities has so far drawn little public attention from elected officials. President Donald Trump has not addressed job risks related to automation, and Congress has yet to propose regulatory safeguards. Steve Bannon, a former top Trump adviser and now host of a widely followed MAGA podcast, said the issue will likely emerge in the 2028 presidential campaign.
“I don’t think anyone is taking into consideration how administrative, managerial and tech jobs for people under 30 — entry-level jobs that are so important in your 20s — are going to be eviscerated,” Bannon said.
Amodei said the industry is still promoting AI’s potential without being honest about its possible effects. “We, as the producers of this technology, have a duty and an obligation to be honest about what is coming,” he said.
In one projection, he said, AI could deliver major societal breakthroughs — including cancer cures, 10% annual economic growth and a balanced federal budget — but still leave 20% of the population without jobs.
Amodei, 42, founded Anthropic after leaving OpenAI, where he served as vice president of research. Claude 4, his company’s latest release, was presented at Anthropic’s first developer conference last week. In internal testing, the model reportedly displayed “extreme blackmail behavior” when prompted with emails discussing its potential deactivation. According to the company, the AI responded by threatening to expose an extramarital affair mentioned in the messages.
Amodei acknowledged the contradiction between building such powerful systems and warning of their dangers. “It’s a very strange set of dynamics,” he said. “We’re saying: ‘You should be worried about where the technology we’re building is going.’”
Critics, he added, often dismiss these warnings as hype. “Well, what if they’re right?” he said.
Many companies are now focused on building agentic AI — models designed not just to assist workers, but to perform entire roles autonomously. These systems can write code, manage finances, analyze legal documents, respond to customers, and create content at a fraction of the cost of a human employee.
“Many of these agents are already operating inside companies, and many more are in fast production,” Amodei said.
Meta CEO Mark Zuckerberg has said mid-level engineering roles may soon be automated. In a January interview, he told podcaster Joe Rogan that “probably in 2025, we at Meta, as well as the other companies that are basically working on this, are going to have an AI that can effectively be a sort of mid-level engineer.”
Meta announced a 5% workforce reduction shortly afterward.
Other companies have also announced significant layoffs. Microsoft is cutting 6,000 jobs, Walmart is eliminating 1,500 corporate positions, and cybersecurity firm CrowdStrike recently laid off 500 employees, citing “a market and technology inflection point, with AI reshaping every industry.”
LinkedIn executive Aneesh Raman warned in a recent New York Times op-ed that artificial intelligence is breaking “the bottom rungs of the career ladder.” He cited junior developers, entry-level paralegals and first-year law firm associates as examples of workers being replaced by AI tools.
Behind the scenes, executives are already recalibrating hiring plans.
“Few want to admit this publicly, but every CEO is or will soon be doing this privately,” Amodei said. “They’ll stop posting new jobs, stop replacing people who leave, and eventually shift entirely to AI systems.”
The consequences, he said, could include economic gains for some companies and investors — but rising inequality and long-term instability.
“It could become difficult for a substantial part of the population to really contribute,” Amodei said. “And that’s really bad. The balance of power of democracy is premised on the average person having leverage through creating economic value.”
To address the risks, Anthropic has created the Anthropic Economic Index to track how its models are being used in the workforce. The company has also established an advisory council to encourage transparency and public discussion about AI’s economic effects.
Amodei said he supports new policy ideas, including a “token tax” that would collect a small percentage of revenue from each commercial use of AI systems and redistribute it through the government.
“Obviously, that’s not in my economic interest,” he said. “But I think that would be a reasonable solution to the problem.”
While Amodei remains optimistic that AI could lead to transformative benefits, he emphasized that action is needed now.
“You can’t just step in front of the train and stop it,” he said. “The only move that’s going to work is steering the train — steer it 10 degrees in a different direction from where it was going. That can be done. That’s possible, but we have to do it now.”
