Center for AI Safety (CAIS) — publication: Representation Engineering: A Top-Down Approach to AI Transparency — proposes methods to read and control LLM internal representations for safety

NoteThe source directly confirms all key elements of the claim: (1) The publication is 'Representation Engineering: A Top-Down Approach to AI Transparency' — exact title match; (2) It is authored by researchers affiliated with Center for AI Safety (Andy Zou, Long Phan, Sarah Chen, Xuwang Yin, Mantas Mazeika, Ann-Kathrin Dombrowski, Shashwat Goel, Nathaniel Li, Zifan Wang, Steven Basart, Dan Hendrycks all listed with CAIS affiliation); (3) The paper proposes methods to 'read and control LLM internal representations' (Section 3 explicitly covers 'Representation Reading' and 'Representation Control'); (4) The methods are framed for safety purposes (abstract mentions 'safety-relevant problems'); (5) The arXiv ID 2310.01405 corresponds to October 2023, matching the 'as of 2023-10' temporal qualifier. All author names in the claim match the source author list.