Mike Vizard summarizes Google’s 2025 DORA survey, examining strong AI adoption in IT, correlations with engineering practices, and key takeaways for DevOps teams and leaders.

Latest DORA Report from Google Surfaces Significant AI Adoption

Author: Mike Vizard

Overview

Google’s annual DevOps Research and Assessment (DORA) survey for 2025 demonstrates significant AI tool adoption among IT professionals. Out of nearly 5,000 surveyed, 90% report using artificial intelligence (AI) tools, with 80% stating they are more productive due to AI.

Key Findings

  • Widespread Use: Median usage of AI tools spans 16.22 months, with IT professionals spending about 2 hours per workday (25%) engaged with AI-driven solutions.
  • AI Engagement: 60% use AI at least half of the time when facing problems or tasks; 7% use it always, while 39% use it occasionally.
  • Trust and Skepticism: Despite adoption, 30% of respondents have little or no trust in AI-generated code, indicating ongoing skepticism.
  • Primary Use Cases:
    • Writing new code (71%)
    • Literature reviews (68%)
    • Modifying code (66%)
    • Proofreading and editing content/images (66%)
    • Analyzing requirements (49%)
    • Internal communications and calendar management (48%, 25%)

Organizational Insights

  • Platform Quality: 90% of organizations utilize at least one internal platform. High-quality platforms directly enhance the ability to extract value from AI, amplifying overall development efficiency.
  • Team Archetypes: Google’s analysis indicates seven team archetypes, from foundationally challenged teams (10%) to harmonious high achievers (20%). More mature teams benefit most from AI’s impact.
  • Performance Spectrum: Metrics for team maturity include lead time, deployment frequency, rework rates, change failure rates, and recovery times.

Strategic Takeaways

  • For leaders, the report stresses investing in platform engineering, automated testing, and DevOps feedback loops as foundations for maximizing AI value.
  • Top-performing teams show that speed and stability are not mutually exclusive; both can be achieved with mature practices.
  • While AI accelerates some tasks (e.g., DevOps workflow automation), its benefits for deeper business logic are less pronounced.

Conclusion

AI adoption is now the norm for DevOps and engineering teams, but realizing its full potential depends on quality engineering practices and robust internal platforms. Organizations that invest in their systems of work will continue to see outsized gains from AI-enabled development.

For further reading, see the 2025 DORA report announcement.

This post appeared first on “DevOps Blog”. Read the entire article here