iamtrask: agent time horizon measures stability, not intelligence — don't conflate ASI signals
iamtrask · x · 2026-11-17
iamtrask pushes back on a popular narrative: an agent's task time horizon is a measure of stability, not intelligence. Stability is necessary but not sufficient for ASI.
Key points:
- A while loop can run forever yet isn't intelligent.
- Today's LLMs have a stability flaw: after a few hours humans must effectively "reboot" them — re-prompt, redirect, etc.
- Researchers worldwide are fixing this flaw, so LLMs will need rebooting less often.
- But that doesn't mean we're approaching superintelligence — models will simply sustain the same level of intelligence for longer stretches without human intervention.
More from AGI Musings
- “Stop aiming for 20% productivity — AI should end work-for-money altogether” — VraserX · 2026-11-17
- Blogger: much of white-collar work was just "button clicking," now over as computers learned English — moonsandhues · 2026-11-17
- Debate: can OpenAI's plan to have agents formally prove safety properties work? — zetalyrae · 2026-11-17
- FT: Firms Investing Heaviest in AI Grew White-Collar Headcount 10.2% — Scobleizer · 2026-11-17
- AI wealth could drive tens to hundreds of billions a year in philanthropy, argues Nan Ransohoff — allisondman · 2026-11-17
- Panagram slop detection could collapse the anti-AI argument, argues MLStreetTalk — burny_tech · 2026-11-17