Anthropic chief: AI is set to eliminate 50% of entry-level positions in the UK

Anthropic chief: AI is set to eliminate 50% of entry-level positions in the UK

Dario Amodei, the chief executive of AI frontrunner Anthropic, has delivered a grave caution regarding the work landscape, forecasting that AI could eliminate as much as 50% of all entry-level white-collar professions in the next five years. The UK market may encounter considerable upheaval during this shift. At Anthropic’s inaugural developer conference, Amodei underscored the…


Dario Amodei, the chief executive of AI frontrunner Anthropic, has delivered a grave caution regarding the work landscape, forecasting that AI could eliminate as much as 50% of all entry-level white-collar professions in the next five years. The UK market may encounter considerable upheaval during this shift.

At Anthropic’s inaugural developer conference, Amodei underscored the rapid adoption of AI, especially sophisticated language models like their premier product, Claude. He stated, “We, as the creators of this technology, have a responsibility and a duty to be straightforward about what lies ahead. I don’t think this is on people’s radar.”

He indicated that this swift advancement might escalate unemployment rates to 10-20%, as companies progressively automate junior positions across industries including law, finance, technology, and consulting.

Impact on UK Employment

While these estimates hold true globally, the UK is already experiencing the repercussions. According to data from the Office for National Statistics (ONS), youth unemployment in the UK climbed to 12.6% in early 2025, alongside stagnant wage increases for entry-level jobs. In technology centers like Cambridge and the City, there has been a significant decline in junior recruitment, partially attributed to the integration of AI.

A report from the venture capital firm SignalFire indicated that hiring for early-career roles in Big Tech has decreased by half since the pandemic, with AI identified as a significant factor. In 2024, just 7% of all new hires were entry-level, a reduction of 25%.

“AI is accomplishing the tasks that interns and recent graduates used to handle,” noted SignalFire partner Heather Dishy. “Now, a single seasoned worker using AI tools can manage the duties of several junior employees without the associated costs.”

Amodei criticized both governments and AI companies for what he perceives as a deficiency in transparency regarding the gravity of the changes ahead. “Most are oblivious to the fact that this is imminent,” he remarked.

In the UK, policymakers are focused on AI safety and national competitiveness, yet Amodei contends that economic preparedness is being neglected. He pointed to deregulations like the EU AI Act and the US’s sweeping “big beautiful bill,” which bans state-level AI legislation for a decade, as potentially endangering workers.

With UK unemployment currently at 4.5% according to the latest data, Amodei cautioned, “We can mitigate the worst of this, but only if we take action immediately.”



  • Insights launches Microsoft Teams personality intelligence integration

    Insights launches Microsoft Teams personality intelligence integration

    Insights launches Microsoft Teams personality insights meeting integration tool today. New Teams app embeds Insights Discovery personality intelligence into everyday workplace meetings.


  • Women in business are building momentum — and business is better for it

    Women in business are building momentum — and business is better for it

    Women in business are gaining ground across the UK economy. Progress in leadership and representation is becoming easier to see, even as obstacles around pay, capital, and care remain. For employers and investors alike, the strongest case for backing women now rests as much in performance as principle.


  • Peers press licensing-first AI training regime

    Peers press licensing-first AI training regime

    Peers urge ministers to reject opt-out AI copyright rules now. A Lords committee says licensed, transparent training data would better support creators, investment, and responsible model development, while warning that weaker copyright protections could stall UK licensing markets and deepen reliance on opaque overseas systems.