Simulating Policy Impacts: Developing a Generative Scenario Writing Method to Evaluate the Perceived Effects of Regulation
This provides a tool for policymakers and researchers to assess and iterate on policies for mitigating AI harms, though it is incremental as it applies existing LLMs to a new domain.
The authors tackled the problem of evaluating policy effectiveness for mitigating AI's negative impacts by developing a method using GPT-4 to generate pre- and post-policy scenarios and translating them into metrics based on human perceptions, finding that EU transparency legislation was perceived effective in areas like labor and well-being but ineffective in social cohesion and security in a user study with 234 participants.
The rapid advancement of AI technologies yields numerous future impacts on individuals and society. Policymakers are tasked to react quickly and establish policies that mitigate those impacts. However, anticipating the effectiveness of policies is a difficult task, as some impacts might only be observable in the future and respective policies might not be applicable to the future development of AI. In this work we develop a method for using large language models (LLMs) to evaluate the efficacy of a given piece of policy at mitigating specified negative impacts. We do so by using GPT-4 to generate scenarios both pre- and post-introduction of policy and translating these vivid stories into metrics based on human perceptions of impacts. We leverage an already established taxonomy of impacts of generative AI in the media environment to generate a set of scenario pairs both mitigated and non-mitigated by the transparency policy in Article 50 of the EU AI Act. We then run a user study (n=234) to evaluate these scenarios across four risk-assessment dimensions: severity, plausibility, magnitude, and specificity to vulnerable populations. We find that this transparency legislation is perceived to be effective at mitigating harms in areas such as labor and well-being, but largely ineffective in areas such as social cohesion and security. Through this case study we demonstrate the efficacy of our method as a tool to iterate on the effectiveness of policy for mitigating various negative impacts. We expect this method to be useful to researchers or other stakeholders who want to brainstorm the potential utility of different pieces of policy or other mitigation strategies.