Broad Timelines — EA Forum
We should have neither short nor long AI timelines, but a very broad probability distribution over when transformative AI will arrive. I explore the implications of this deep uncertainty.
Ryan Greenblatt 4mo 40 6 2 I agree with many particular points in this post and the apparent thesis [1] , but also think most people [2] should focus on short timelines (contrary to the apparent implication of the post). The reasons why are: Short timelines have more leverage. This isn't just because of more neglectedness now, but also because: (1) it's easier to target approaches towards shorter timelines where less has changed, (2) short timelines are riskier (and I think riskier worlds are more leveraged for most interventions, this is sensitive to my views on risk and the most leveraged in
related reading
- Short AI Timelines Aren’t Always Higher-Leverageforethought.org
- Broad Timelines - Toby Ordtobyord.com
- AI Timelines — LessWronglesswrong.com
- The case for multi-decade AI timelinesepochai.substack.com
- AGI and the EMH: markets are not expecting aligned or unaligned AI in the next 30 years — Basil Halperinbasilhalperin.com
- "Long" timelines to advanced AI have gotten crazy shorthelentoner.substack.com
- Short Timelines Don't Devalue Long Horizon Research — LessWronglesswrong.com
- My picture of the present in AI — LessWronglesswrong.com
- Consider chilling out in 2028 — LessWronglesswrong.com
- AI Timelines: Where the Arguments, and the "Experts," Standcold-takes.com
- Two-year update on my personal AI timelines — LessWronglesswrong.com
- A (Slightly) Mechanistic Theory for Exponentially Increasing AI Time Horizons? — LessWronglesswrong.com