AI magnifies your teams strengths - and weaknesses, Google report finds (www.zdnet.com)

Google’s 2025 DORA (DevOps Research & Assessment) report — based on a 5,000‑respondent survey and 100+ hours of interviews — finds AI is now mainstream in software development and acts as an amplifier: it speeds up output but magnifies both strengths and dysfunctions. Key stats: 90–95% of developers use AI (a 14% year‑over‑year jump), median interaction is ~2 hours/day, 80% report higher productivity but only 59% report improved code quality, and 70% trust AI’s outputs. The study maps eight team and delivery factors (e.g., team performance, throughput, instability, burnout) into seven archetypes, showing top teams can be both fast and stable while weaker teams suffer amplified failure modes. The report’s technical and organizational implications are clear: successful AI adoption is a systems problem, not just a tools problem. Seven practices drive results — AI policy, strong data ecosystems and accessible data, rigorous version control, small batch changes, user focus, and shared quality platforms — and two enablers flip AI from liability to multiplier: robust internal platform engineering (adopted by ~90% of orgs) and mature value‑stream management (VSM). Practically, teams should treat AI adoption as an organizational transformation: invest in platforms, data pipelines, VSM and core engineering disciplines to turn AI’s productivity gains into reliable, high‑quality delivery.