Scott Alexander, curated
← Back to curation

Metaculus Monday 2/8/21

Quality
63
Strong
Claude Shift
42
Moderate
RWI
1
of 10

Summary

An early Metaculus Monday. Two threads give it standalone weight: (1) a real prediction-market-infrastructure report — Polymarket's mispriced 6%-Trump-by-March contract traced to brutal USDC/relayer on-ramp friction, and the genuinely significant news that Kalshi got CFTC approval to run a regulated event-futures exchange with $25k caps (the onion-futures carve-out gag); (2) a walkthrough of Metaculus's AI questions — the AGI-by-2036 definition and curve shape, the human-machine-parity Turing bet, and the hard-vs-slow-takeoff GDP-doubling questions (will we see doubling-every-4-years before doubling-every-1-year), with Scott's own skepticism about a GDP discontinuity. Informative but a curated tour rather than original analysis.

Why this score

Quality 63 · Strong. Solid-plus / low-Strong (63). Above the bare-roundup floor thanks to the Kalshi regulatory milestone and the substantive takeoff-speed walkthrough, but it is fundamentally market-surveying with light commentary.

Claude’s paradigm shift 42 · Moderate. Moderate (42). Reports others' questions/markets; the takeoff-discontinuity skepticism is a Scott aside, not a new frame.

Real-world impact 1 · Negligible. 1 — a forecasting roundup; negligible real-world effect.