In May 2025, Anthropic CEO Dario Amodei told Axios that artificial intelligence could eliminate half of all entry-level white-collar jobs and push unemployment rates to between 10 and 20 percent within one to five years. This stark prediction has reframed the ongoing debate about AI's impact on employment, shifting focus from future hypotheticals to imminent disruptions.
The Core Warning
Amodei's forecast is based on the rapid advancement of large language models and generative AI systems, which are increasingly capable of performing tasks traditionally reserved for junior professionals. He emphasized that the timeline for these changes is not decades away but could materialize within the next half-decade. The prediction suggests that roles in fields such as customer service, data entry, legal research, and basic accounting are particularly vulnerable.
According to Amodei, the potential unemployment surge to 10-20% would represent a dramatic increase from current levels in most developed economies. For context, the U.S. unemployment rate in early 2025 stood at approximately 3.8%, while the Eurozone hovered around 6.5%. A jump to double digits would mark the highest unemployment rates since the Great Depression in many regions.
Economic and Social Implications
Amodei's warning has sparked debate among economists, policymakers, and technologists. Some argue that AI will also create new job categories, as previous technological revolutions have done. However, the speed and scale of AI adoption could outpace the ability of workers to reskill. The CEO noted that entry-level positions have traditionally served as a training ground for career advancement, and their elimination could disrupt the entire career ladder for young professionals.
"We are looking at a scenario where the bottom rung of the career ladder is pulled away," Amodei said in the Axios interview. "This isn't just about job loss; it's about the loss of pathways to develop skills and experience."
Policy Responses and Reskilling
Governments and corporations are beginning to respond. Several countries have launched AI task forces to study labor market impacts, and proposals for universal basic income (UBI) have gained renewed attention. However, Amodei cautioned that policy responses must be swift and substantial. He advocated for expanded reskilling programs, portable benefits, and potentially a job guarantee for displaced workers.
Tech companies, including Anthropic, face pressure to implement responsible AI deployment. Anthropic has positioned itself as a safety-focused AI firm, but even its CEO acknowledges that the technology's benefits come with significant risks. "We are building powerful tools, and with that power comes responsibility," Amodei said. "We need to think hard about how to distribute the gains and cushion the losses."
Historical Context and Future Outlook
Historical parallels, such as the Industrial Revolution and the rise of the internet, show that technology can ultimately lead to higher productivity and new jobs. Yet the pace of change in AI is unprecedented. The McKinsey Global Institute estimates that by 2030, up to 30% of work activities could be automated, but Amodei's timeline is more aggressive.
Critics argue that Amodei's prediction may be overly pessimistic, noting that AI still struggles with complex reasoning, creativity, and human interaction. However, the rapid improvement of models like Anthropic's Claude suggests that these limitations may be temporary. The debate now centers on whether society can adapt quickly enough to prevent widespread disruption.
In the meantime, educators and employers are rethinking the skills needed for the future workforce. Emphasis is shifting toward critical thinking, emotional intelligence, and technical skills in AI development and oversight. The message from Amodei is clear: the AI jobs debate is no longer about if, but when and how severe the impact will be.
