Developers are worried that they're living on borrowed time as AI continues creeping into their daily workflows, according to new research.
In a survey of 1,000 developers across the UK, US, Germany, Spain and the Netherlands, Adaptavist found that nearly half believe that software development is becoming a less attractive career because of AI, and three-in-ten said AI is already negatively affecting their career progression.
Respondents cited a lack of hands-on learning opportunities, a squeeze on mid-career developers, a shift from writing code to managing AI agents, and an ongoing ‘verification tax’.
Indeed, developers said AI is changing job functions to such an extent that they now spend more time reviewing and verifying code than ever before.
The impact of AI on junior developers
A key finding from the Adaptavist report centers around overreliance on AI tools, with seven-in-ten worried that junior developers are becoming too reliant on the technology.
This, the study noted, raises serious concerns about whether future generations will develop foundational skills needed to progress in the field. This isn’t the first time these concerns have been raised in the industry.
As reported last year, tech blogger and programmer Namanyay Goel warned that the influx of AI tools could undermine basic coding skills in years to come.
Sign up today and you will receive a free copy of our Future Focus 2026 report - the leading resource for IT decision-maker insight on priorities and investment areas in AI, security and more.
Overall, 63% of respondents told Adaptavist that AI has reduced the amount they learn through hands-on coding, and 44% believe AI is making core software engineering skills less important.
Crucially, 43% said they feel they have fewer learning opportunities because of AI - and this rises to 45% among experienced developers.
Matt Saunders, DevOps lead at Adaptavist, said the findings show a fundamental shift in the profession, and warned against a creeping trend of overreliance.
"AI is changing what it means to be a software developer, but organizations cannot afford to lose sight of how engineers develop their skillsets,” he commented.
“Developers may be relying on automation for more of their work, but they still need hands-on experience to understand the code, solve problems and make sound technical decisions.”
Mid-level devs are getting anxious
According to Adaptavist, AI is affecting developers in very different ways depending on what stage they’re at in their career.
For mid-career developers, many believe AI is negatively affecting career progression, compared with just 26% of early-career developers and 27% of “experienced developers”.
With this in mind, nearly half (44%) of mid-career devs would get rid of AI in the workplace if given the choice. A concerningly high number (31%) of early-career devs would also do the same.
Saunders said that while organizations need to give junior developers the opportunity to build “strong foundations”, support for more experienced engineers is also needed as their roles evolve.
“Organizations need to help developers grow alongside the technology, ensuring they have the skillset and support to use Ai effectively throughout their careers,” he said.
Devs are stuck bot-sitting
In terms of agent management, nearly seven-in-ten developers said monitoring and overseeing agents has now become an explicit part of their daily role. Around 82%, for example, regularly deploy multiple agents for any given task.
With this influx, seven-in-ten believe managing AI agents is set to become a core software engineering skill.
As previously reported, ‘bot-sitting’ has become a recurring pain point for workers across a range of professions in recent months - and it’s having a marked impact on productivity.
Findings from a study conducted by Glean’s Work AI Institute found that although 90% of workers use AI in their daily activities, productivity gains are being offset as they fight to make the technology usable.
For developers, Adaptavist pointed to a similar dynamic at play. Many are cautious about AI-generated work following recent high-profile incidents, with four-in-ten now reviewing AI-generated content more closely.
Nearly one-third (29%) said the technology requires tighter human oversight, and 21% are less willing to give agents full autonomy.