Economist Novosad: Pangram scores can't convict AI writing, but help triage suspicion
paulnovosad · x · 2026-10-23
In his debate with Obhishek Saha, Paul Novosad clarifies that a 100% Pangram score cannot prove someone secretly used AI to write, say, a referee report — so it should not be used as a filter or grounds for shaming.
Still, he argues that when a text "smells way off" in both content and style, a Pangram check is a useful verification signal. Attention is limited, he says, so he relies on the signals at his disposal, errors notwithstanding.
More from AGI Musings
- Economist on AI detector Pangram: not a filter, but a useful verification signal — paulnovosad · 2026-10-23
- Beff Jezos: regulation won't stop internet agents, capability diffusion is safer — beffjezos · 2026-10-23
- AI Now's People's AI Assembly spotlights data center opposition and labor rights — AINowInstitute · 2026-10-23
- VraserX: build an abundance civilization, don't romanticize austerity — VraserX · 2026-10-23
- Kuprel: AI Ends Long Hours but Rewards Precision Decisions at the Right Moment — Kuprel · 2026-10-23
- After data and compute, the next picks-and-shovels are talent and physical infrastructure — philhchen · 2026-10-23