Four weeks and 10B tokens: AI agents take a crack at the 1960s deletion channel problem
DimitrisPapail · x · 2026-09-17
Dimitris Papail published a long-form account of tasking AI agents with the deletion channel capacity problem, open since the 1960s.
- Roughly four weeks of GPU time and over 10 billion tokens, mixing classic literature with agent reasoning
- The agents did not resolve the problem, but the write-up offers a candid case study of agentic research and where current models stall
- He argues the failure itself is informative, mapping the gap between hype and actual open-math capability
The article is being shared as a rare first-hand log of AI's real level in scientific research.
More from AGI Musings
- Alpha School investigation: 17 campuses on probation, Bill Ackman as shadow lead, Holmes involved — benjaminjriley · 2026-11-02
- Google's chief economist: AI will transform more jobs than it kills — gains come from redesigning workflows — danielrock · 2026-11-02
- Daniel Lemire: AI is exploding demand for boring compute, server CPUs to double — lemire · 2026-11-02
- Alignment researcher: the word "aligned" implies consistencies that don't exist — davidmanheim · 2026-11-02
- By 2035, Indian Parents Will Ask How Many Agents You Manage, Not Your Salary — theteknosaur · 2026-11-02
- Under 5% of US consumers pay for AI, yet compute demand is already forcing usage cuts — NinaDSchick · 2026-11-02