Half of developers worry AI is reshaping their careers

A new Adaptavist survey of 1,000 developers across five European and U.S. markets finds AI is changing how software engineers learn, work and advance. The findings point to rising career anxiety, especially among mid-career developers, as teams shift from coding to supervising AI tools and agents.

Why it matters: - AI is no longer just a productivity tool for developers. The research suggests it is changing career paths, skill-building and the future value of software engineering roles. - More than half of developers, 54%, say they are anxious about their long-term careers as AI changes the skills they need and the work they do. - 46% say software development is becoming a less attractive career because of AI. - Almost a third, 31%, say AI is already hurting their career progression.

What happened: - Adaptavist published the New Developer Realities whitepaper today. - The study surveyed 1,000 developers in the UK, US, Germany, Spain and the Netherlands. - The research found that AI affects developers differently depending on career stage. - The findings show AI is changing both the developer role and how developers view software engineering as a career.

The details: - Developers pointed to fewer hands-on learning opportunities, pressure on mid-career roles, a move from writing code to managing AI agents and a growing “verification tax” from reviewing AI-generated work. - Seven in 10 developers, 70%, worry that junior developers are becoming too dependent on automation. - 63% say AI has reduced how much they learn through hands-on coding. - 44% say AI is making core software engineering skills less important. - 43% say they have fewer learning opportunities because of AI. - That figure rises to 45% among experienced developers. - More than a third of mid-career developers, 35%, say AI is hurting their career progression. - That compares with 26% of early-career developers and 27% of experienced developers. - 44% of mid-career developers say they would remove AI from the workplace if they could. - That compares with 31% of early-career developers and 40% of experienced developers. - More than half of developers, 51%, say they are managing AI rather than simply using it. - 64% describe AI as a teammate. - 47% say AI is actively replacing parts of their role. - 68% say managing AI agents is becoming an explicit part of their daily work. - 82% say they regularly deploy multiple AI agents for one task, while 16% use a single agent. - 70% believe managing AI agents will become a core software engineering skill. - 66% trust AI agents to complete tasks autonomously. - 60% trust AI agents to make important technical decisions. - After high-profile incidents involving AI systems and agents, including the Hugging Face cyber incident, 39% say they review AI-generated work more closely. - 29% say AI output now requires tighter human oversight. - 21% say they are less willing to give AI agents autonomy. - 20% say they are more cautious about adopting new agents.

Between the lines: - The research points to a role shift, not just a tooling change. Developers appear to be moving from building code directly to supervising, verifying and coordinating AI output. - That shift helps explain why anxiety is highest among mid-career developers, who may be expected to adapt quickly without losing the depth of experience built through hands-on work. - The findings also suggest a widening gap between AI adoption and workforce training, especially for junior developers who need foundational practice. - Matt Saunders, DevOps Lead at The Adaptavist Group, said organizations should not lose sight of how engineers build skills as automation takes on more work.

What's next: - Adaptavist says employers will need to give junior developers more opportunities to build fundamentals. - The company also argues experienced engineers will need support as their jobs expand into managing and overseeing AI agents. - The research suggests software teams will keep adjusting oversight practices as AI adoption grows and trust remains uneven.

The bottom line: - AI is changing developer jobs fast, but the bigger risk may be how much it is reshaping career development, not just day-to-day coding.