YouTube is a $60 billion annual business, bigger than Netflix's $45 billion, with 2 billion users, 13% annual growth, and $100 billion paid to creators over the past four years. CEO Neal Mohan sat down with Big Technology's Alex Kantrowitz to explain how AI fits into that machine without breaking it. The core argument: AI improves the infrastructure around content, not the content judgment itself.
Mohan draws a clear line. AI can test thumbnails, personalize recommendations, and answer viewer questions mid-video. It cannot, in his framing, decide what is good. 'Nobody gets to tell us on YouTube what's good, what's not good, what's art, what's not art,' he said. He also named AI slop directly as a threat to YouTube Shorts, which puts the platform in the uncomfortable position of scaling AI tools while policing their worst outputs. The interview surfaces real tension between algorithmic optimization and creative autonomy, and Mohan does not fully resolve it.
The full transcript is worth reading for three specific threads: how YouTube's collaboration with Google DeepMind is shaping product decisions, whether AI-generated answers to viewer questions will cannibalize creator views, and what Netflix chasing podcasters means for YouTube's dominance in long-form audio-video. Mohan's answers are careful in ways that reveal exactly where the pressure is.
[READ ORIGINAL →]