Why AI hasn’t replaced software engineers, and won’t
Simon Willison · Simon Willison · 2026-06-14
Arvind Narayanan and Sayash Kapoor, summarized by Simon Willison, argue that AI has not caused mass software engineering layoffs because the true bottlenecks — deciding what to build, verifying outputs, and applying deep contextual human understanding — remain resistant to automation even as AI accelerates code writing.
Appears in
Extraction
Topics: ai-job-displacementsoftware-engineeringai-automation-limitsfuture-of-work
Claims
- Empirical data does not support the narrative that AI is causing mass unemployment in software engineering.
- In New York's first full year of AI-disclosure WARN Act filings, not a single company among 160+ filers cited AI as a cause of layoffs.
- The real bottlenecks in software engineering are deciding what to build, verifying and being accountable for deliverables, and deep human understanding of the codebase and business context.
- AI accelerates the code-writing phase, but code writing is not the primary bottleneck in software development.
- Because software engineering has few regulatory barriers and high AI suitability, its resilience to AI-driven layoffs suggests most other professions will be even more protected.
Key quotes
In March 2025, New York became the first U.S. state to add an AI disclosure checkbox to WARN Act filings. In the full first year, more than 160 companies filed WARN notices. Not a single one checked the AI box.
The real bottlenecks (1) deciding and specifying what to build, (2) verifying and being accountable for what is delivered, and (3) the deep human understanding — of the codebase, the business, and the environment — required to carry out both of these.
Give me all of the AI assistance in the world and the value I produce will still be reliant on how deeply I understand both the problems and the solutions that the agents are building for them.