Co·Insights

Co·Insights was designed based on ethnographic research with Asian American and Pacific Islander (AAPI) communities. We conducted more than 40 interviews with AAPI community organizations, fact-checkers, academics, and platforms. The Co·Insights consortium includes world-leading academics at the University of Massachusetts Amherst, Rutgers University, the University of Connecticut, the University of Pennsylvania, and the University of Colorado Boulder.

Disinformation targeting AAPI communities is too often out-of-sight from mainstream misinformation response efforts: it is in languages and on platforms beyond not traditionally considered in the United States. Co·Insights brings together community organizations, academics, and fact-checkers to address misleading and harmful narratives. In this way we're fostering interracial solidarities, greater understanding, and better democracy.

Our overarching goal is to empower people to understand and evaluate the information they see online. We do this by building tools such as "misinformation tiplines" that people can choose to consult if they want to learn more about a given piece of content. All of our tools are opt-in, and we're building our technology in close collaboration with AAPI communities who are underserved currently due to the diverse languages, platforms, and communities involved. We are working to promote a healthy and inclusive public sphere online in which everyone—including historically marginalized and minoritized groups—can equally participate.

We seek to reimagine and move beyond the current fact-checking paradigm. Current fact-checking efforts tend to focus on one claim or piece of content at a time. In contrast, we are identifying more persistent, long-lasting narrative themes and patterns that are broader than a single claim. Our opt-in tools allow individuals to seek out additional context when they choose. We're supporting community-centered responses beyond misinformation by facilitating cross-cultural dialogue, interracial solidarities, and educational materials. This approach can move the status quo away from content moderation and towards an approach more akin to email spam detection or local antivirus software.