Owain Evans
Source-listed: Director and Research Lead · Truthful AI
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- TruthfulQA: Measuring How Models Mimic Human FalsehoodsSaved credit: Contributor · 2021Explore people connected to this work →
- The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"Saved credit: Contributor · 2023Explore people connected to this work →
- Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMsSaved credit: Contributor · 2024Explore people connected to this work →
- Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMsSaved credit: Contributor · 2025Explore people connected to this work →
- Subliminal Learning: Language models transmit behavioral traits via hidden signals in dataSaved credit: Contributor · 2025Explore people connected to this work →