Why Humans Can't Keep Up With AI-Written Code - Tariq Shaukat, Sonar

Agents have made code cheap: one study saw lines of code up 2,450% but releases up only 30%. Sonar CEO Tariq Shaukat draws on the factory revolution that Toyota and Lexus started to explain how to build agentic software factories you can trust. He breaks down five hard problems: human review that doesn't scale (people agree with AI that's wrong 4 out of 5 times), agents struggling with large codebases, errors compounding over long tasks, unreliable LLM judges, and near-zero time to exploit. Then he lays out Sonar's code confidence formula: intent, guide rails, verification and evidence. Chapters: 0:00 Intro 0:26 The ball bearing that changed car factories 2:23 Software already borrowed these ideas 2:55 Agents break the factory 5:19 Problem 1: Human in the loop doesn't scale 6:45 Problem 2: Agents struggle with big codebases 8:04 Problem 3: Errors compound 9:06 Problem 4: LLM judges are unreliable 10:03 Problem 5: Time to exploit is near zero 10:25 The code confidence formula 12:36 Results from customers 15:49 Code is abundant, confidence is scarce Tariq Shaukat https://www.sonarsource.com Join us in SF for the second annual AIE CODE summit: https://aie.codes our last flagship summit of 2026. Applications live now!