BairesDev survey results link wider use of AI tools in software development with higher perceived career control among women engineers, alongside ongoing concerns about fairness and responsibility for AI-driven outcomes.
Its quarterly Dev Barometer found that 64% of women respondents said AI improved their ability to shape their career path, while 84% reported career momentum tied to growth in AI skills.
The findings come with caution on trust. Only 56% of women developers said they trust AI tools to be fair across genders. The research did not examine why respondents perceived fairness gaps.
Accountability also emerged as AI becomes more common in day-to-day development work. Across all respondents, 51% said responsibility for AI-generated outcomes falls on them personally, while another 11% said accountability remains unclear.
The results suggest responsibility is shifting inside engineering organisations. AI tools can generate code, tests, documentation, and other outputs, but many developers still expect to carry the burden for what reaches production, even when AI systems contribute to the work.
"AI isn't a side tool anymore. It's embedded in production workflows. And once it touches production, accountability changes. That's great for career mobility, and infrastructure demands accountability," said Nacho De Marco, CEO and co-founder of BairesDev.
He also pointed to a stronger focus on checking AI outputs inside teams. "As adoption accelerates, our data shows developers feel personally responsible for AI outcomes. What we're seeing is a shift toward disciplined review. The teams that scale AI successfully are the ones that treat validation as engineering, not as an afterthought," he said.
Comfort and caution
Respondents reported broad comfort with AI tooling. Overall, 88% said they feel comfortable using AI tools, and 58% said the benefits outweigh the overhead.
Even with that level of familiarity, developers highlighted verification as a skills priority. A total of 56% ranked critical evaluation of AI output as the baseline skill for 2026, and 67% said teams lack sufficient knowledge to validate AI-generated results.
Time pressure remains a barrier to checks. One in five respondents (20%) cited deadline pressure as the primary barrier to verification, pointing to a gap between the aspiration of thorough review and the reality of delivery schedules, particularly in teams adopting AI across production workflows.
What accountability means
BairesDev described AI accountability in operational terms, including code review and validation, testing standards, documentation requirements, and production monitoring for enterprise deployments.
"Building accountability into AI-assisted processes doesn't have to be complicated," De Marco said. "AI is drafting the first version of the code, but a human still owns the final decision. That means validating outputs, testing rigorously, and taking responsibility before anything goes live."
Access and culture
The survey also examined what developers think determines who benefits most from AI adoption at work. The top factor was access to tools, training, and infrastructure (67%), followed by willingness to experiment and adapt (63%). Seniority and experience came next (50%).
Among women respondents, the ranking shifted. Team culture and norms ranked above seniority as determinants of advancement through AI skills: 49% cited team culture and norms, compared with 38% who pointed to seniority.
The results suggest that both technical access and the organisational environment shape how quickly people can translate AI skills into progression. Training budgets, approved tools, and internal guidance can determine who gets hands-on practice, while team expectations can determine whether experimentation is rewarded or treated as a risk.
Survey details
The Q1 2026 Dev Barometer surveyed 1,329 developers across 61 countries. Women made up 25% of the sample, and 47% of respondents reported having eight or more years of professional experience.
BairesDev said it will continue tracking how AI adoption affects distributed engineering teams as AI tools become more embedded in production workflows.