A generative renaissance: Shutterstock’s AI transformation
Recently, the record label of pop stars Drake and the Weeknd forced streaming and social media platforms to remove an AI-generated song that replicated the artists’ style.
For Michael Francello, Shutterstock’s Director of Innovation, the move symbolized the need for a better understanding of the value of AI to the artistic process. “AI is additive,” he says. “It will never replace the beauty and complexity of human creativity.” Brands have a responsibility to the creators they represent to make sure both sides can benefit.
That’s the approach Francello and the team at Shutterstock are taking toward AI-generated images. The company recently rolled out an AI image generator, allowing users to create at the speed of their imaginations. One of the important things that makes this image generator different from others is that Shutterstock is paying their contributors whose work was used in training the model through their Contributor Fund. Artists also get a cut when a user licenses an image that leverages their work. Francello says he likes his current role because it puts him in a position to fight on behalf of artists at a time when AI technology is going to impact them, one way or another.
The discourse around generative AI and art typically has sparked debate regarding its impact on artists. Artists can’t make money when AI tools can produce content of similar quality at much greater velocity and volume, the thinking goes. That may be true in a world where AI tools indiscriminately scrape the web for training data with no regard for copyright or authorship. But Shutterstock points to a different path, one where artists are brought along for the ride and can claim their rightful share of the value being created.
People are always going to build new technologies that disrupt the incumbent business side of artistic pursuits, Francello says. He cites the fight Metallica had with Napster in the early 2000s in which the band went to court against the music sharing service to protect its copyright. This stirred controversy and earned Metallica some ill will from fans, but Francello says the band never should have had to take that action. It should have been the record label fighting to protect its client’s intellectual property. Maybe the label could have worked out some kind of licensing deal for its band. And who knows, maybe this could have sparked innovation in the business model that earns bands and labels money.
We’re at a similar moment today when it comes to generative AI. Many artists and creative agencies are trying to block their work from being leveraged in models. This makes sense when companies are not paying artists whose work contributed to its model. But simply saying no may not be the best option in the long term. AI technologies are here today. They’re already creating new ways of bringing media into the world, which is creating an opportunity for artists and, in particular, the agencies that represent them to create innovative new ways of participating in the AI economy. So, the question is not so much whether you’re going to be a part of it. The better question may be: What part do you want to play?
- Ed Burns, NExT Futurist and Writer