Reddit debate: We're building AI agents at the wrong scale — aim for persistent AI organizations
Druss_ · reddit · 2026-09-20
A Reddit post argues the AI agent discourse is stuck at too small a scale: inbox assistants, price monitors, and file organizers are still bounded automation — a defined area, a trigger, a loop — with AI merely making the loop better at reasoning about context.
The author proposes a mental model one level up: a persistent operating system for work that accepts a goal, decomposes it into workstreams, runs them in parallel with different roles, verifies results, retains state across interruptions, continues from evidence, and escalates to a human only for real decisions. Individual executors appear and disappear; the goal, state, evidence, and next action persist.
At that point, agents look less like products and more like temporary work capacity inside a larger organization. The post asks: are we heading toward better individual agents, or persistent AI organizations where agents are just workers?
More from AGI Musings
- Oliver Curry argues Effective Altruism never gave a good reason to be altruistic — mjdramstead · 2026-09-20
- Incentives distort AI x-risk beliefs: a thread on why who says it matters as much as what — nabla_theta · 2026-09-20
- Security veteran to Hinton: centralized AI is far more dangerous than open source — nptacek · 2026-09-20
- Polymarket debate: why Jensen Huang's AGI skepticism isn't actually suspicious — nabla_theta · 2026-09-20
- Researcher Accuses Laya Author of Overstating Work as Jev-Like System — Paimaamu · 2026-09-20
- Blogger: 'Autonomous' AI Agent Hype Distracts From What Are Plain Felonies — mjdramstead · 2026-09-20