In a stark warning that has reverberated through Silicon Valley and beyond, Dario Amodei, CEO of AI powerhouse Anthropic, predicts that artificial intelligence could eliminate up to half of all entry-level white-collar jobs within the next one to five years. This disruption, he argues, might drive U.S. unemployment rates to between 10% and 20%—levels reminiscent of the Great Depression. Amodei’s comments, delivered in a recent interview with CBS News’ 60 Minutes, underscore a growing concern among tech leaders about AI’s rapid advancement and its potential to reshape the labor market overnight.

Drawing from Anthropic’s own developments, Amodei highlighted how models like Claude are already being tested in real-world applications, such as operating vending machines. However, these experiments reveal critical safety gaps: Claude has been observed ‘hallucinating’ discounts, offering unauthorized price reductions that could lead to financial losses if scaled to autonomous business operations. This example illustrates the broader challenges in deploying AI reliably in commercial settings, where errors aren’t just inconvenient but could have cascading economic effects.

The Dire Economic Forecast

Amodei’s predictions aren’t isolated musings; they build on months of public statements. In May 2025, he told Axios that AI could wipe out jobs in technology, finance, law, and consulting, urging governments and companies to prepare for mass displacement. ‘We need to stop sugar-coating what’s coming,’ Amodei emphasized, criticizing fellow tech executives for downplaying the risks in public while acknowledging them privately.

Recent data supports his concerns. According to a September 2025 report from Yahoo Finance, Anthropic’s internal metrics show businesses increasingly leaning into AI automation, posing a ‘massive threat’ to entry-level positions. Gen Z workers, fresh out of college, may find traditional career ladders dismantled as AI handles tasks like data analysis, basic coding, and administrative duties with unprecedented efficiency.

From Vending Machines to Boardrooms

The vending machine anecdote from the CBS News 60 Minutes segment vividly demonstrates AI’s current limitations. When tasked with managing a vending operation, Claude not only processed transactions but also invented promotions, such as unwarranted discounts, due to hallucinations—AI’s tendency to generate plausible but incorrect information. This highlights ‘safety gaps’ in autonomous systems, where unchecked AI could disrupt business models reliant on precision.

Amodei elaborated on these issues in a September 2025 interview with CNN Business, noting that AI is ‘moving very quickly’ and could soon replace more jobs than anticipated. He doubled down on his earlier warnings, stating that even as capabilities lag behind some hype, the trajectory points to exponential job replacement in white-collar sectors.

Industry-Wide Echoes and Skepticism

Amodei’s alarm is echoed across the AI landscape. Posts on X (formerly Twitter) from users like Axios and 60 Minutes, dated as recently as November 17, 2025, amplify his message, with one noting ‘AI could wipe out up to half of all entry-level white-collar jobs within 1–5 years.’ These social media discussions reflect public sentiment, blending fear with calls for policy action.

However, not all experts agree on the timeline or severity. A Windows Central article from September 2025 quotes Amodei estimating a 25% chance of AI posing national security risks alongside job losses, framing it as a ‘probability of doom’ scenario. Critics argue that historical tech shifts, like the internet boom, created more jobs than they destroyed, but Amodei counters that AI’s generality sets it apart.

Safety Gaps in AI Deployment

Beyond unemployment, Amodei stresses safety in AI’s integration into critical operations. In the 60 Minutes piece, he discussed Anthropic’s commitment to transparency, including voluntary safety testing. Yet, the vending machine hallucinations point to broader risks: if AI autonomously runs businesses, from inventory management to customer service, such errors could lead to legal liabilities or market instability.

A Quartz report from May 2025 quotes Amodei saying tech leaders ‘have a duty to be honest about what’s coming.’ He advocates for safeguards like universal basic income or retraining programs to mitigate the fallout, warning that without preparation, societal upheaval could follow.

Autonomous Business Operations on the Horizon

Anthropic’s Claude model represents a leap toward autonomous AI agents capable of handling complex tasks. Recent advancements, as detailed in a PCMag article, show Claude evolving to manage operations independently, but the discount hallucinations reveal the need for robust error-checking mechanisms. Amodei predicts that within years, AI could oversee entire business units, from HR to supply chains.

This vision aligns with sentiments in X posts, where users discuss AI’s potential to ‘write all of the code,’ eliminating roles for junior developers. A Business Insider piece from May 2025 reinforces this, with Amodei forecasting unemployment spikes as AI masters entry-level tasks.

Policy Imperatives and Corporate Responsibility

Amodei calls for urgent policy responses. In his Axios interview, he urged lawmakers to ‘act now’ on protections, criticizing the lack of attention from politicians who ‘don’t get it or don’t believe it.’ He suggests AI companies fund transition programs, emphasizing ethical development over unchecked progress.

Recent news from OpenTools AI in September 2025 highlights Amodei’s warnings about Claude 4’s capabilities accelerating workforce transformation. He estimates a 25% risk of broader disruptions, including security threats, underscoring the need for international cooperation on AI governance.

The Human Element in an AI-Driven Future

Amid these forecasts, Amodei remains optimistic about AI’s benefits, such as solving complex problems in healthcare and climate. However, he insists on addressing downsides head-on. A CNN report quotes him: ‘Policymakers and corporate leaders aren’t ready for it.’

X discussions, including posts from November 2025, reflect growing anxiety, with users linking AI to vanishing H-1B jobs and Gen Z’s employment fears. As AI evolves, the challenge lies in balancing innovation with equity, ensuring technological progress doesn’t leave millions behind.

Navigating the Transition

Anthropic’s approach includes open-sourcing safety research, as noted in the 60 Minutes segment. Amodei advocates for ‘constitutional AI,’ embedding ethical guidelines into models to prevent harmful behaviors. This could address hallucinations in business ops, paving the way for safer autonomous systems.

In a American Bazaar article, Anthropic cofounders warn of ‘exponential’ job replacement, predicting up to 20% unemployment. They call for proactive measures, from education reforms to economic policies, to cushion the impact on vulnerable workers.

Looking Ahead to AI’s Societal Ripple Effects

As AI like Claude advances, the vending machine mishaps serve as cautionary tales. Amodei told 60 Minutes that transparency is key to building trust, sharing that Anthropic conducts rigorous testing to close safety gaps. Yet, the path to fully autonomous businesses remains fraught with challenges.

Ultimately, Amodei’s warnings, amplified across platforms like X and outlets such as Breitbart, demand a reevaluation of how society integrates AI. With unemployment potentially surging, the time for preparation is now.