The Future of Generative AI & Online Knowledge Communities
Thursday, November 12th & Friday, November 13th
Frances Searle Building, Center for Human-Computer Interaction + Design, 2240 Campus Drive
How are generative AI systems interacting with knowledge commons and the open internet? What is the current landscape of threats and opportunities in light of recent developments? What strategic initiatives and research are needed to advance the public interest in sustaining public digital infrastructure and knowledge resources?
A few years into the adoption of commercial LLMs, an adversarial, destructive relationship has emerged between generative AI systems and the online communities that create and sustain critical knowledge commons. Wikipedia, Reddit, StackExchange, OpenStreetMap, free/libre and open source software (FLOSS) projects, and more have created thriving ecosystems of public participation and invaluable digital public infrastructure. These resources have also provided critical inputs to large language models (LLMs) and other generative AI technologies. Increasingly, these same AI technologies threaten the public interest in sustaining knowledge commons and the open internet. Communities face an expanding set of AI-driven challenges at unprecedented scale: aggressive scraping, disingenuous or destructive engagement, disintermediation, and lost contributions. Without concerted and coordinated action, the current trajectory risks sending these communities and shared knowledge commons into a downward spiral.
The Community Data Science Collective, with support from the Alfred P. Sloan Foundation and the Northwestern Center for Human-Computer Interaction + Design, will convene a dialogue to propose a public interest research agenda to sustain knowledge commons and digital public infrastructure in an age of AI. We will build upon an earlier meeting convened at the Bellagio Center in 2024 that proposed a research agenda on these topics. Two years later, we seek to understand what has changed, what questions have been answered, and—more urgently than ever—what public interest technologists and researchers can do to advance a socially beneficial equilibrium for the future.
Event details for participants
Back to top