Screens were filled and images were filled before discussions about the future of music and AI between YouTube/Google Music Chiefs Lyor Cohen and Rapper-Singer Jidenna.
This is the official music video for Wu-Tang Clan's latest single, “Mindingo,” all of its photorealistic frames were made with Google AI video generator Veo 2.
“When I saw that, I almost fell out of my chair,” Cohen said at the event Monday night. “I challenge anyone to say, 'It's not art.' It's absolutely artistic and incredibly well done. ”
Zidenna spent the next 30 minutes detailing her evolution from a music artist to an AI-savvy multi-system, first pausing and greeted Busta Rhymes, Chuck D, and Wyclef. The hip-hop legend was one of a crowd of dozens in a screening room at the branch of New York's recently opened private club, San Vicente.
Large-scale social follow and hits like the Grammy-nominated 2015 track “Classic Man” appeared during Jidenna's major label days. Today, he is clearly creating independent courses, throwing himself into the booming realm of computer-generated videos.
Zidena maintained that she could draw a direct line between hip-hop rises and AI rises. “We're at a point when our people, who may have been marginalized for some reason, take the tools and think about what they can do with it,” he said. “And like hip hop, they say it's not art. Like hip hop, they say, 'Oh, it's illegal. You can't do it. The model is trained with other people's data.' That's what we call sampling. ”
While Google has a clear incentive to promote AI, the conversation was notable for not grabbing the toes on many worries about the technology that plagues the entertainment business. There was approval of copyright protection, the dangers of deepfakes, and the need for an evolved business model, but the general message was the “come to yes” variety.
“There's no way to stop this from happening,” Cohen said. “So my idea is to get on our front legs and embrace this moment. Let's believe in the fact that truly talented and creative people can take it to the next level.”
Zidenna said it's important to plunge into the unknown. “You have to fear that leap,” he said. And for many of us, it's a very scary moment. But if you accept it, trust me, I learned everything through trial and error.
To provide legal context, Cohen sat in the audience with Kevin Montler, a senior corporate lawyer for YouTube music. Montler said he was asked to take his views on the current state of AI. “This technology is ahead of the law. And it's not that continuity is seen in position, even within industry groups, both on the technological and right-wing sides. I think everyone is trying to understand that.”
Zidenna addressed her own views on legal uncertainty, confirming that artists should not use the work to train AI models without their consent, or that they would create fraudulent images based on their likeness on video generators. Still, he agreed to Monler's point and noted that he attended Stanford University at the same time as Openry co-founder Sam Altman. “I don't think we have a system that's fast enough to keep up with high tech,” he said.
In response to accusations of AI images that look “fake,” he said, “is not more fake than the people who post happy days in their lives. Social media is more fake than me. But they call it artificial.”
The speed at which AI is evolving was another theme in the lecture. Cohen asked how often the tools change, and Zidenna replied, “Every week.” Living in LA gives him a connection in the film industry, and he says industry workers call the first day of the week a “disastrous Monday.” “I call it 'magical Monday'. “He smiled. “If you don't embrace change, it's going to be miserable.”
