Wiener's 1960 Alignment Warning Revisited: Patterning as a Control Problem
geoffreyirving · x · 2026-11-11
The quoted post notes that Norbert Wiener, forefather of control theory and arguably the first alignment researcher, warned in 1960 that we had better be sure "the purpose put into the machine is the purpose which we really desire."
The author uses this to frame a new post reading "patterning" (in the Toy Model of Superposition) as a control problem, linking classic cybernetics to modern interpretability research.
More from AGI Musings
- RethinkX shuts down research arm after ten years, framing it as a graduation — adam_dorr · 2026-11-11
- Naval: Vibe Coding Is the New Product Management, Model Tuning Is the New Coding — aakashgupta · 2026-11-11
- Kai-Fu Lee: next wave of AI value creation shifts from models to industry applications — kaifulee · 2026-11-11
- Delegating to AI makes price signals more valuable—and crypto more compelling — curious_vii · 2026-11-11
- 13 LLMs with 320 past experiments in context all beat baseline at predicting research payoff — cihangxie · 2026-11-11
- CAIS warns AI nears automating all remote work as Claude completes 41.3% of real freelance gigs — Polymarket · 2026-11-11