Dev proposes rules for AI email agents: disclose identity, separate inbox, keep humans cc'd
curious_vii · x · 2026-10-19
The author argues that any AI agent continuing an email conversation on a user's behalf should clearly disclose that it's a computer, operate from a separate inbox, and keep the user copied on every message.
The core ask is transparency and oversight: users should be able to see what their agent is doing at all times, rather than delegating communication to a black box — a pointed take on identity disclosure as agents start acting on our behalf.
More from AGI Musings
- DHH vs François Chollet: On Inventing the Future vs Believing It Into Existence — manosaie · 2026-10-19
- Investor Stewart Alsop: You Are in a Recursively Self-Improving Dyad With Your AI — StewartalsopIII · 2026-10-19
- Autoresearch democratizes getting results, not doing better research — a PhD student's warning — manosaie · 2026-10-19
- Data centers use just 0.85% of global primary energy, by the numbers — DarinFeinstein · 2026-10-19
- LLM optimizer tuning is planning, argues rebuttal to 'planning is overrated' — HanchungLee · 2026-10-19
- ChatGPT conversation yields a sharp line: who profits from others' beliefs while externalizing the cost of being wrong — sebpaquet · 2026-10-19