New paper proposes three autonomy levels for software development
dawnsongtweets · x · 2026-07-28
Inspired by SAE levels for self-driving, the authors propose three autonomy levels for software development, defined by which SDLC stages AI fully owns across design, implementation, testing, deployment and demand. The goal is to give the field a precise vocabulary for responsibility and failure modes as AI takes on more of the pipeline.
More from coding & agent
- Coding agent runs are getting long — one dev suggests breathing, not parallel sessions — teropa · 2026-10-16
- Dev builds MCP layer so Claude Code can work on servers without ever holding SSH keys — NETMESAFE · 2026-10-16
- As coding agents run longer autonomously, one dev's tip: don't parallel-session, just breathe — teropa · 2026-10-16
- 'Refusing to code with AI in 2026 is like refusing Google in 2005' — prasenx · 2026-10-16
- Claude Code adds idle compaction to save 1h prompt cache on idle sessions — EricBuess · 2026-10-16
- Handling agents that crash mid-task: the hard part is the unknown outcome across the network — Responsible-Dare5139 · 2026-10-16