Credit: Samuel Axon

AI tools are widely used by software developers, but those devs and their managers are still grappling with figuring out how exactly to best put the tools to use, with growing pains emerging along the way.

That's the takeaway from the latest survey of 49,000 professional developers by community and information hub StackOverflow, which itself has been heavily impacted by the addition of large language models (LLMs) to developer workflows.

The survey found that four in five developers use AI tools in their workflow in 2025—a portion that has been rapidly growing in recent years. That said, "trust in the accuracy of AI has fallen from 40 percent in previous years to just 29 percent this year."

The disparity between those two metrics illustrates the evolving and complex impact of AI tools like GitHub Copilot or Cursor on the profession. There's relatively little debate among developers that the tools are or ought to be useful, but people are still figuring out what the best applications (and limits) are.

When asked what their top frustration with AI tools was, 45 percent of respondents said they struggled with "AI solutions that are almost right, but not quite"—the single largest reported problem. That's because unlike outputs that are clearly wrong, these can introduce insidious bugs or other problems that are difficult to immediately identify and relatively time-consuming to troubleshoot, especially for junior developers who approached the work with a false sense of confidence thanks to their reliance on AI.

As a result, more than a third of the developers in the survey "report that some of their visits to Stack Overflow are a result of AI-related issues." That is to say, code suggestions they accepted from an LLM-based tool introduced problems they then had to turn to other people to solve.

Even as major improvements have recently come via reasoning-optimized models, that close-but-not-quite unreliability is unlikely to ever vanish completely; it's endemic to the very nature of how the predictive technology works.

That's why 72 percent of the survey participants said that "vibe coding" is not part of their professional work; some feel it's too unreliable, and it can introduce hard-to-debug issues that are not appropriate for production.

Why devs use the tools anyway

So given all that skepticism and frustration, why are devs still using the tools? Well, in some cases, their managers are trying to force them to. But more commonly, it's because the tools are still clearly useful—it's just important not to misapply them.

It's important that managers and individual contributors bring AI tools into the workflow alongside robust training to ensure a deep understanding of best practices so the tools aren't misused in a way that creates more problems than it solves or that wastes more time than it saves.

Developers need to be less trusting of things like Copilot autocomplete suggestions, treating them more as a starting point rather than just hitting tab and moving on. Tools like that are best suited for a sort of limited pair programming relationship: asking the LLM to find problems or suggest more elegant solutions that you take into critical consideration, not to suggest complete methods that you take at face value.

They can also be useful for learning. The opportunity to always be learning by continually building familiarity with new languages, frameworks, or methodologies is one of the things that draws some people to the job, and LLMs can reduce friction in that process by answering questions in a more targeted way than is possible with laborious searches through often incomplete technical documentation—exactly the sort of thing that people have historically used StackOverflow for in the past.

"Although we have seen a decline in traffic, in no way is it as dramatic as some would indicate," StackOverflow Chief Product and Technology Officer Jody Bailey said in a comment to VentureBeat. StackOverflow plans to commit some of its resources both to expanding AI tool literacy and to fostering community discussions that help solve issues that are specific to workflows that involve those tools.