Owain Evans proposes a practitioner guide to subliminal learning and phantom transfer
OwainEvans_UK · x · 2026-11-09
Following his RSI data-poisoning direction, Owain Evans proposes a practical project: reviewing current findings on subliminal learning, phantom transfer, and related subtle-generalization phenomena, and spelling out their practical implications for AI practitioners. Ideally effects would be tested in production-like setups, though a review alone makes progress. AI agents can assist with the review and brainstorming, but human involvement is essential. The deliverable should be a maximally accessible, succinct document — not exhaustive, but covering the key content needed to understand and evidence the practical implications.
More from Safety
- Nvidia board opens independent investigation into Jensen Huang over Chinese spy link — tomwarren · 2026-11-09
- South Korea unveils $610B budget for 2027, nearly doubling AI spending — Dr_Singularity · 2026-11-09
- Boaz Barak shares full AI safety course videos, from Schulman to Altman — eigenron · 2026-11-09
- Dev's satirical hack idea turns Meta smart glasses' surveillance back on their wearers — kscottz · 2026-11-09
- Gas generators for AI data centers spark concerns over noise and NOx pollution — bigdata · 2026-11-09
- Owain Evans proposes a research direction: data poisoning risks during RSI — OwainEvans_UK · 2026-11-09