Simon Willison discusses the November 2026 inflection point where AI models like GPT-5.1 and Claude Opus 4.5 crossed a threshold in code generation reliability, enabling practical coding agents that can build functional applications. The conversation covers implications for software engineering workflows, testing bottlenecks, and the emergence of 'dark factories' with increased automation.
Background
AI code generation has been rapidly evolving, with models like GPT and Claude continuously improving their ability to understand and generate programming code. The discussion focuses on when these tools became reliably useful for practical software development.
- Source
- Simon Willison
- Published
- Apr 3, 2026 at 04:40 AM
- Score
- 7.0 / 10