Over 21 researchers from Google DeepMind, Harvard, and other institutions have published a white paper arguing that images and video could be as important as text for developing artificial general intelligence. The paper, titled Visual General Intelligence: A White Paper and published on arXiv as 2608.25924, outlines a research agenda for what the authors call visual general intelligence, or VGI. Contributors include Robert Geirhos from Google DeepMind and Yilun Du from Harvard. The researchers argue that generative video models and self-supervised learning could provide a foundation for AGI that language alone cannot. The work builds on DeepMinds previous publications, including Levels of AGI (2024) and From AGI to ASI (2026).
Source: Read the original article
