In context: Artists and content creators continue to become more entwined with artificial intelligence. While some view AI as an infringement on the work of others, some creators see it as a new medium to present their ideas. YouTube creator and a cappella singer Peter Hollens clearly sees it as an opportunity. The social media star recently released a music video that was 100% created using AI technologies.

What happens when you give creative license to an AI and ask it to generate a music video? Apparently, nothing quite as bad as we've come to expect.

YouTube celebrity and a cappella singer Peter Hollens recently proved that it can be something more promising than a computer simply presenting users with unwanted romantic advances or alerting them to their demise. Hollens recently posted several tweets giving fans a glimpse into a 100% AI created music video to accompany his song I Can't Make You Love Me.

According to Hollens, several AI tools were used together to create the music video. ChatGPT was used to generate a shotlist that described what visuals would be used in each scene of the video. Images were generated from text using Midjourney, exported, and later animated using kaiberAi.

Hollens' likeness in the video, seen singing in several sequences, was also created by AI.

D-ID, an AI tool used to create animated avatars, photos, and videos, allows users to create anything from fantasy-based avatars to talking portraits generated from a single image. Additional editing and revisions were completed using AI-based tools from RunwayMI.

It's important to note that the singing, lyrics, and other audio in the video are not AI-generated, at least not this time.

Vocal credit goes solely to Hollens, a well established a cappella singer who has been creating YouTube content for more than a decade. With more than 2.8 million subscribers, Hollens is no stranger to creating highly produced and well received audio and video content.

That's not to say that AI isn't capable of creating its own music. OpenAI's Jukebox is a neural network capable of generating music and singing across any number of genres and styles. No musical talent, production skills, or vocal training is required. Select a genre, select an artist to use as a base pattern, and feed in some lyrics. Jukebox will take it from there, and the results aren't always as bad as you'd expect.

The results will only get better as the technology matures, eventually catching up to reality TV in proving that sometimes you don't need talent to be famous.