Google just dropped its annual DORA report, and the numbers are staggering. AI adoption among software developers has hit 90% - a 14% jump from last year - with professionals now spending two hours daily working with AI tools. The research, surveying nearly 5,000 tech professionals globally, confirms what many suspected: AI has become an essential part of the developer toolkit, fundamentally changing how software gets built.
Google Cloud's latest DORA report just landed, and it's painting a picture of an industry in rapid transformation. The 2025 State of AI-assisted Software Development study surveyed nearly 5,000 technology professionals worldwide, revealing that AI has moved from experimental tool to daily necessity for the vast majority of developers.
The headline number? AI adoption has rocketed to 90% among software development professionals, marking a substantial 14% increase from the previous year. But it's not just adoption - it's integration. These developers, product managers, and tech professionals are dedicating a median of two hours daily to working with AI tools, weaving them into their core workflows.
The productivity gains are undeniable. More than 80% of respondents say AI has enhanced their productivity, while 59% report it's positively influenced their code quality. Google's research shows that 65% of those surveyed are heavily relying on AI for software development, with 37% reporting "moderate" reliance, 20% "a lot," and 8% "a great deal."
But here's where it gets interesting - and complicated. Despite these widespread benefits, Google uncovered what they're calling a "trust paradox." While developers are using AI extensively, only 24% report having "a great deal" or "a lot" of trust in these tools. Meanwhile, 30% trust AI only "a little" or "not at all."
"This indicates that AI outputs are perceived as useful and valuable despite a lack of complete trust in them," the report notes. It suggests AI is functioning as a supportive enhancement rather than a replacement for human judgment - a nuanced relationship that's defining this transition period.
The organizational impact tells a more complex story. While AI is clearly boosting individual performance, its effects on teams and companies vary dramatically. This year's research shows AI adoption is now linked to higher software delivery throughput - teams are shipping more code and applications, reversing last year's concerning findings. However, ensuring that software actually works as intended before reaching users remains an ongoing challenge.
Google's research identified what they call AI acting as both "mirror and multiplier." In cohesive organizations with strong processes, AI amplifies efficiency. In fragmented teams with poor workflows, it highlights and potentially magnifies existing weaknesses.
