As deepfake videos become increasingly realistic and more common online, figuring out how we can identify them is an important challenge. But are humans or the leading AI models currently more capable of detecting deepfake videos? And how should content moderation systems be designed to detect and flag deepfakes or other forms of manipulated media?
In this session, leading researchers from MIT and Johns Hopkins will share their pioneering research on how humans and AI compare at deepfake detection, including how we can combine the strengths of both, and what this could mean for content moderation and fighting disinformation.
