We take an initial look at where median forecasts are (mostly) underestimating and (sometimes) overestimating AI progress, adoption, and impact.

2 points•gmays•4 days ago•2 comments•

2 comments

michaelastreiko4 days ago
The calibration angle is useful: separating capability forecasts from deployment and adoption forecasts makes misses easier to diagnose.
katamin4 days ago
I think this will widen the gap even more.

Read the full thread on Hacker News →

Related stories