AI coding tools: near-universal adoption, uneven corporate support
GitHub’s second annual developer survey—this time covering 2,000 respondents across the U.S., Brazil, Germany, and India—paints a clear picture: individual developers have embraced generative AI, but their employers have not all caught up. While software engineers made up most of the sample, the survey also included data scientists and software designers to capture a broader view of AI’s impact across disciplines.
The headline number is striking: more than 97% of respondents in every country said they have used AI coding tools at work at some point. Yet organizational support varies widely by region. In the U.S., 88% of respondents said their companies either actively encourage or allow AI tool use. In Germany, that figure drops to 59%.
Company stance on AI adoption
When asked to characterize their employer’s approach to AI coding tools, respondents split into two camps. Across all markets, 30–40% said their organizations actively promote adoption. Another 29–49% said their companies allow the tools but offer limited encouragement. That leaves a substantial minority in every region whose organizations have taken no formal position—or are actively discouraging use.
The corporate stance appears to correlate with developer experience. Among respondents at organizations that actively promote AI tools, 48% described their toolchains as “simple” to use. At organizations with a neutral stance, that number jumps to 65% describing their toolchains as complex. The survey suggests AI coding tools may be simplifying workflows—but only where companies have embraced them.
Reported benefits: code quality, onboarding, and time savings
Respondents identified several concrete benefits from AI coding tools, starting with code quality. Perceived improvements were strongest in the U.S. (90%) and India (81%), with Brazil (61%) and Germany (60%) reporting lower but still majority-positive responses.
AI tools also appear to lower the barrier to unfamiliar codebases. Between 60% and 71% of respondents across countries said these tools make it “easy” to adopt a new programming language or understand existing code. A further 23–29% called the effect “very easy.”
Test case generation is another widespread use case. More than 98% of respondents said their organizations have experimented with AI for generating tests, and most use it at least sometimes. That practice is most common in the U.S. (92%) and least common in Germany (65%).
The time savings from these tools are being reinvested in higher-level work. In the U.S. and Germany, 47% of respondents said they use the extra time for collaboration and system design—a continuation of the trend GitHub first observed in last year’s survey of U.S. developers.
Expectations for AI’s future impact
Respondents are broadly optimistic that AI coding tools will help them meet customer requirements. Optimism ranges from 61% in Germany to 73% in the U.S. Notably, confidence tracks with corporate support: developers at companies that actively encourage AI use are more likely to believe the tools drive customer satisfaction.
Expectations for security improvements are nearly universal, with 99–100% of respondents anticipating that AI will enhance code security. Indian developers hold the strongest view, with 41% expecting significant improvement.
Proficiency with AI coding tools is also becoming a career asset. Nearly all respondents (99–100%) believe this skill makes them more attractive to employers, and 43% in Germany and 56% in India say it significantly boosts their employability.
What the survey means for engineering leaders
Three takeaways stand out from the data. First, generative AI is no longer an emerging trend—nearly every respondent has tried these tools. Second, developers are using the time saved on routine work for more strategic activities like collaboration, system design, and learning. Third, organizational policy lags individual behavior: a meaningful share of companies still neither encourage nor formally allow AI tool use.
That gap matters because the benefits of AI coding tools appear to compound where organizations provide active support. The survey suggests that companies with clear guidelines, measurable outcomes, and a deliberate adoption strategy are better positioned to capture the productivity and quality gains developers are already experiencing on their own.
