China AI Digest (May 10, 2026)

Three vocabularies, one quietly unifying question: when AI is no longer a tool you reach for but a layer threaded through the social graph, the code base, and the imagination, what are platforms, labs, and artists actually betting on? AI Weekly Talk closes its Magnificent Seven series with Meta — arguing Zuckerberg's real bet is not on AI but on the next social network, and that whoever owns the entry points wins the decade. Zhang Xiaoyun's two-and-a-quarter-hour conversation with Ohio State's Professor Su Yu walks through six decades of agent research and names the bottleneck that resets the field — continual learning. ChinaTalk hosts Ken Liu — translator of the Three-Body Problem and the Dao De Jing — for a Daoist reading of AI, including the line of the episode: the deeper risk is not machines becoming human but humans being trained to behave like machines.