Beyond The Hype: How Generative AI Could Help Democracy

Generative AI will disrupt politics as usual. Could that be a good thing?

The following is a guest post by Aaron Spitler, an analyst specializing in digital technologies, human rights, and democratic governance.

ChatGPT has prompted much speculation about how generative artificial intelligence (AI) will impact society. From healthcare to academia, experts have maintained that this novel technology will upturn sectors of all types. However, much of the debate surrounding generative AI’s benefits and drawbacks has centered on what it will do to democracies—potential harms run the gamut, from supercharging disinformation campaigns to eroding media trust.

While these concerns deserve our attention, this technology may provide advantages that could strengthen democracy. Improving civic education, remaking legislative processes, and even promoting participatory governance are promising opportunities that may be made possible through generative AI. Should guardrails be put into place, this innovation could create systems of governance that are responsive to the people.

Raising Voter Awareness 

Keeping voters informed about the issues that affect them has been floated as a way generative AI could rejuvenate democracies. In an article published by The Conversation, researchers at Purdue University highlighted how this technology could ensure individuals know how to stay civically engaged. The authors noted how difficult it can be for most people to understand complex policy debates, let alone grasp where their elected representatives stand on matters. To address this problem, the researchers looked at how AI-enabled tools, like chatbots, could make a difference.

AI chatbots could also interactively help people think through complex issues, learn new skills and determine their policy stances, while also providing relevant news stories and facts.

These cutting-edge solutions, powered by large language models (LLMs) trained on reliable data, would answer common public policy questions. Critically, the LLMs’ responses could be tailored to those without subject matter expertise. Citizens, in turn, would be equipped with the knowledge to more actively participate in civic life. In leveraging AI-powered tools like these, the dynamic between constituents and their officials may become more constructive.

Enhancing Citizen Engagement

Voters are not the only ones who could be empowered by generative AI; policymakers could also leverage the technology to become more responsive to those they represent. A piece published by the MIT Technology Review shed light on how generative AI could be used when creating “microlegislation,” referring to aspects of policies or laws that cater to specific groups or interests. In short, the technology would be tasked with identifying and modifying certain parts of a bill to incorporate feedback from constituents, a responsibility that sits within the wheelhouse of generative AI. 

It [the work of microlegislation] is a natural starting point for AI because it’s tightly scoped, involving small changes guided by a clear directive associated with a well-defined purpose.

The value brought forth by this innovation comes at a critical time. Recent polling from the Pew Research Center suggests that many Americans are highly dissatisfied with what they perceive as a dysfunctional, if not broken, system of government. If used ethically and responsibly, this application of generative AI could ensure statutes are more responsive to the public’s needs, restoring citizens’ faith in their government.

Creating Deliberative Policymaking

Finally, generative AI could make policymaking more collaborative and democratic. In fact, this technology has already been used to ensure individuals are meaningfully involved in making decisions. Take the example of Polis, an AI-powered platform that collects and analyzes user responses to policy-related questions. After gathering this feedback, the platform identifies areas of consensus among the submissions it receives. Findings are then shared with users, which later could be used to craft solutions around mutual interests.

Ideally, outputs from this collaboration could also be provided to government leaders, starting a dialogue that might not have existed previously. This formula for revitalizing the dynamic between citizens and their representatives has been successful in the past, as Polis was used in Taiwan to craft consensus-based regulations for Uber back in 2019. Lessons from this case study underline how generative AI-based tools have real potential to breathe new life into democratic processes.

AI’s relationship with democracies is not zero-sum. Malign actors using AI to diminish people’s faith in their institutions, for instance, is just one alarming example. However, when used ethically and appropriately, generative AI could be a force for good in democracies. It could keep voters engaged and informed in ways other technologies fail to accomplish. It may revolutionize how policymakers draft legislation, ensuring constituents' perspectives are not overlooked. It may even bring officials and citizens closer together to craft solutions to the problems that matter most.

Aaron Spitler is an analyst specializing in digital technologies, human rights, and democratic governance. The views expressed here are entirely his own. He can be reached at aaron.m.spitler@gmail.com.