AI exposure is highest in skilled roles like programming and finance, not low-wage jobs, Anthropic data shows

Most people have long believed that automation would first affect lower-skilled, lower-paid roles. The common assumption was that factory work, delivery services and routine manual jobs would be the earliest to disappear.

However, Anthropic’s March 2026 Labor Market Impacts Report points to a very different reality. The professionals currently most exposed to artificial intelligence are those with higher education, significant experience and above-average earnings, the report shows.

The findings in this report turn conventional thinking upside down. There’s a clear tension here: the qualifications and expertise that once helped secure a stable career may now be the very factors placing workers at greater risk of disruption.

“The workers most exposed to AI today are not the least skilled; they are often the most educated and highest paid,” said Bengaluru-based Gagana Sri B, an AI labour market analyst.

These are far from entry-level positions. They are the kinds of jobs people train for, spend years building careers around, and typically consider secure because they depend on specialized knowledge and expertise.

“About 75% of programming tasks are already being handled by AI systems in real-world work settings,” said Ijin M Biju, a technology analyst tracking enterprise AI adoption, based out of Bengaluru.

It is important to clarify what “observed exposure” means in this context, as this is not a forward-looking assumption about what AI could potentially achieve a decade from now. Observed exposure measures how artificial intelligence is currently being deployed to complete tasks in real professional settings. That difference matters. The divide between what AI is capable of and what it is already doing in the workplace has essentially disappeared.

The labour market is already factoring in this transition. According to the report, every 10 percentage-point increase in a role’s AI exposure is linked to a 0.6 percentage-point decline in projected job growth through 2034, based on US Bureau of Labor Statistics data. This is not hypothetical alarmism. Official employment projections are already being revised downward for roles where AI is increasingly performing core tasks.

“For every 10 percentage point increase in AI exposure, projected job growth declines by 0.6 percentage points through 2034,” says Delhi-based Anjana Rajagopalan, a labour economist, citing employment projections.

The entry point into these careers is tightening. Companies aren’t laying off experienced employees at scale, but they are clearly bringing in fewer entry-level hires for positions where AI can now take on a substantial share of the workload. The pipeline for new talent is shrinking from the ground up.

“Hiring of workers aged 22 to 25 into AI-exposed roles has dropped by around 14% since ChatGPT’s launch,” said Karthik K, a recruitment consultant working with early-career hiring data in Bengaluru.

For mid-career professionals, this signal is especially important. It offers a clear indication of where employer demand is moving. The labour market is actively reshaping itself. Jobs that are likely to expand are those that involve guiding AI, assessing complex outputs and tackling nuanced challenges that machines aren’t equipped to handle. On the other hand, roles that are beginning to weaken are those where the primary contribution has been routine task execution.

The Nexford Alumni Outcomes Report 2025 offers a clear snapshot of what success looks like today. The data shows that 82% of alumni report an increase in salary after graduating. One in three experienced salary growth of 50% or more. In addition, within 18 months of completing their degree, 54% had transitioned into management or leadership positions. Right after graduation, 96% were part of the workforce, 86% employed, and 10% self-employed, while only 3% were unemployed.

The individuals behind these outcomes share a defining characteristic: they are fluent in working with data, capable of managing automated systems, able to lead teams through rapid change, and skilled at making strategic decisions that machines cannot replicate. This ultimately comes down to curriculum and relevance, not just the cost of education. In 2026, a degree delivers value only if it reflects how work actually operates today.

Choosing to upskill isn’t about reacting out of fear. It’s about acting at the right moment. Those who move when these shifts are still early indicators, rather than waiting until they become full-scale disruptions, are far more likely to stay ahead than struggle to catch up later.

“Ask yourself a direct question: does your current skill set enable you to guide and leverage AI tools, or does it leave you vulnerable to being replaced by them? Reflect on what your previous education actually prepared you for in a data-driven, AI-integrated business environment. Would it still be relevant in today’s market?” said Rashmi, counsellor at Sahyadri College, Mangaluru (Mangalore), Karnataka.

The takeaway is that large-scale displacement is not the current story, but the labour market is already reorganizing itself around AI. Waiting for unemployment figures to spike may mean reacting too late.

Skills such as business strategy, data literacy, people management and applied AI capabilities are consistently emerging as areas employers prioritize. The Nexford Alumni Outcomes Report 2025 highlights that 54% of graduates move into management or leadership roles within 18 months—positions that operate above the task layer AI is increasingly automating.

The key signal to assess is whether the curriculum reflects how businesses function today and whether graduates demonstrate measurable improvements in performance and earnings. Nexford’s 2025 alumni data showing 82% salary growth post-graduation suggests that relevance is the deciding factor.