Developer: I let AI auto-approve in low-stakes apps, but read every PR for my core products
feregri_no · x · 2026-11-05
Developer feregrino describes a tiered stance on AI automation: he happily lets AI review and approve changes in low-stakes apps, but for the products at the core of his livelihood he reads every single PR — refusing what he calls "cognitive surrender." A concise articulation of the risk-based trust boundary many engineers are drawing around AI coding agents.
More from AGI Musings
- Senior-operator access with intern-level oversight: the failure built in on day one — mclynd · 2026-11-05
- Veteran operator on AI agents: the hard question isn't model smarts, it's delegated authority — mclynd · 2026-11-05
- Ilya: with personal AI agents, humans will no longer be participants — danfaggella · 2026-11-05
- Gradual Disempowerment: AI replacing human roles may erode institutional accountability — ryan_t_lowe · 2026-11-05
- Paper asks whether LLMs are rational: coherence norms published in AI journal — peterbhase · 2026-11-05
- Researchers clash over whether aligning AI to human goals wrongs the AI itself — rgblong · 2026-11-05