Philosophically-motivated questions

Analysis by Charles Rathkopf Last updated: June 2026

[Questions to be written]

Abstract

As large-scale language models increasingly impact safety-critical domains, ensuring their reliable adherence to well-defined principles remains a fundamental challenge. We introduce Deliberative Alignment, a new paradigm that directly teaches the model safety specifications and trains it to explicitly recall and accurately reason over the specifications before answering. We used this approach to align OpenAI’s o-series models, and achieved highly precise adherence to OpenAI’s safety policies, without requiring human-written chain-of-thoughts or answers. Deliberative Alignment pushes the Pareto frontier by simultaneously increasing robustness to jailbreaks while decreasing overrefusal rates, and also improves out-of-distribution generalization. We demonstrate that reasoning over explicitly specified policies enables more scalable, trustworthy, and interpretable alignment.

Citation for this analysis

Charles Rathkopf, “Philosophical Questions in Deliberative Alignment: Reasoning Enables Safer Language Models,” AI Deception Papers, June 2026, https://doi.org/10.48550/arXiv.2412.16339