PolicyPad: Collaborative Prototyping of LLM Policies
This work addresses the problem of involving domain experts in LLM policy prototyping for fields like mental health and law, offering a tool to improve participatory AI alignment, though it is incremental in applying UX methods to this context.
The paper tackled the challenge of collaborative policy design for LLMs in high-stakes domains by developing PolicyPad, an interactive system that facilitates rapid experimentation and feedback, resulting in enhanced collaboration and novel policy contributions in workshops with domain experts.
As LLMs gain adoption in high-stakes domains like mental health, domain experts are increasingly consulted to provide input into policies governing their behavior. From an observation of 19 policymaking workshops with 9 experts over 15 weeks, we identified opportunities to better support rapid experimentation, feedback, and iteration for collaborative policy design processes. We present PolicyPad, an interactive system that facilitates the emerging practice of LLM policy prototyping by drawing from established UX prototyping practices, including heuristic evaluation and storyboarding. Using PolicyPad, policy designers can collaborate on drafting a policy in real time while independently testing policy-informed model behavior with usage scenarios. We evaluate PolicyPad through workshops with 8 groups of 22 domain experts in mental health and law, finding that PolicyPad enhanced collaborative dynamics during policy design, enabled tight feedback loops, and led to novel policy contributions. Overall, our work paves participatory paths for advancing AI alignment and safety.