Notification tragedy of the commons: a personal agent should suppress ~95% of them
signulll · x · 2026-09-16
Each assistant optimizes locally for helpfulness, so they all independently decide that surfacing one more thing is useful — globally you get a notification tragedy of the commons. The actual personal agent probably has to do the opposite: ingest notifications from everything, suppress 95% of them, rank the remainder against your current context, and decide when not to interrupt you. A sharp take on personal agent design.
More from AGI Musings
- Ex-OpenAI researcher: Vatican could uniquely broker AI risk agreements — jachiam0 · 2026-09-16
- A retrocausal strong anthropic principle argument for why AI doom never happened — eigenhector · 2026-09-16
- Ex-OpenAI staffer: AI ends humanity not with malice but by consuming our resources — AndyMasley · 2026-09-16
- After a year of agentic coding, dev concedes traditional artists were right: no soul to be found — ryunuck · 2026-09-16
- Zuckerberg on AI slowdown debate: alignment is becoming the key capability, not a brake — eleiber · 2026-09-16
- cyantist: Humans are secularly ignorant but spiritually wise, AIs the reverse — ZeroStateReflex · 2026-09-16