I’ve been sitting on this one for a bit, but the latest episode of Google’s Dialogues on Technology and Society series is worth a watch. They sat LL COOL J down with James Manyika, Google’s SVP of Research, Technology & Society, to talk AI and creativity.
If you’re expecting a dry corporate chat, you’ll be surprised. LL COOL J brings the same energy he brings to a cypher, and Manyika is clearly a fan who also knows his stuff. The conversation moves from how AI can help musicians find new sounds to the more existential question: does it make you less of an artist if a machine helps write the hook?

What I found interesting is that LL didn’t dismiss AI outright. He talked about using it as a tool for inspiration—like a digital jam session partner that never gets tired. He gave a concrete example: feeding a beat into a model and having it suggest alternative basslines or chord progressions. That’s not cheating, that’s collaborating with the tech.
Manyika, for his part, kept it grounded. He pointed out that AI in music isn’t new—auto-tune and digital samplers were once seen as threats too. The difference now is the scale and speed. A model can generate a hundred variations of a melody in seconds. But as he said, the hard part—the taste, the timing, the emotional hook—still belongs to the human.
I’ll admit, I rolled my eyes a bit when they started talking about “democratizing creativity.” That phrase gets thrown around way too much. But LL made a good point: not everyone can afford a producer or a studio. If AI can help a kid in a bedroom make something that sounds professional, that’s a win. He also warned that the industry needs to protect artists’ rights, especially around voice cloning and style mimicry. Fair point.
The episode doesn’t pretend AI is a magic wand. LL talked about how he still writes lyrics by hand, on paper, in the middle of the night. The machine can help with structure, but it can’t replace the grind. That’s the kind of honesty I appreciate.
If you’re curious about where AI and music are headed, this is a solid 40 minutes. No hype, just two people who actually make things talking about how tools change the process. And LL COOL J still sounds like he’d destroy you in a freestyle, even if the beat was generated by a transformer model.
Comments (0)
Login Log in to comment.
Be the first to comment!