Madonna’s a Material Girl Living (Performing) in a GenAI World

Longtime Madonna collaborator Sasha Kasiuha says generative AI was key to creating visuals to accompany two songs during the Material Girl’s recently completed World Celebration Tour.  

Kasiuha, who served as the tour’s content director, says the stage visuals for both “Isla Bonita” and “Take a Bow” were “entirely generated” using Runway’s tools, according to a Q&A shared on the Runway website.

He said Madonna has “always been open to embracing new technologies that can unlock new visual elements.”    

GenAI was not their first stop, however. Kasiuha explained, “We originally tried traditional CGI, but it looked pretty flat and cheesy.” He says he has long incorporated AI into his creative process and turned to it for these tasks because “it made sense” and because Midjourney and Runway “helped us achieve our specific vision for the tour.”

Kasiuha said, “‘La Isla Bonita’ is a song that transports the audience to an ethereal, tropical place, and we wanted the visuals to match that type of heavenly vibe.” 

First, they turned to “text to video prompts like ‘surreal sunset, clouds’ to get a quick variety of options and to help us land on a stylistic direction.” Once they settled on the project’s style, they utilized Runway “Gen-1 to transfer styles onto existing videos.” However, he notes, these videos featured “water, tunnels, animals flying, or nebulas — things that had really great movement and contrast.” That content was then transformed again via a cloud “reference image.” 

Ultimately, Kasiuha said they achieved  something that wasn’t “quite real” but also wasn’t “quite CGI.” Instead, the videos “were surreal and had a quality that’s totally different and hard to describe — that’s what we were going for.”

Kasiuha noted that the production was achieved by a team of six in a few months — “quite quick” for a tour of that scale, he said.

They debuted this backdrop content first, and they said audiences had an “overwhelmingly positive” reaction, bolstering them “to continue experimenting with new techniques,” he said. 

For their second AI project, the 29 year-old Ukrainian native said, “The creative vision for ‘Take a Bow’ included more of a narrative with a cohesive theme — we wanted to bring the audience to a world beyond the screen.” 

As with the first project, they began with Midjourney’s text-to-image prompts, aiming to create “a world inspired by the surreal realms depicted by Dali and Magritte,” which Kasuiha calls “ an exploration of infinity.” These photos were then input in Runway Gen-2 to animate them.

“The camera and motion controls that exist in Gen-2 were indispensable to creating these visuals in a way that was consistent with our specific vision,” he explained. 

READ MORE: Generating new worlds for Madonna’s Celebration Tour with Runway (Runway)

“[S]ome AI video enthusiasts say the emerging technology could one day upend entertainment, enabling you to choose your own movie with customizable story lines and endings,” the AP’s Matt O’Brien writes in his coverage of the tour’s innovation.

He describes Madonna as one of the “early adopters,” noting she has “long pushed art’s boundaries.”

READ MORE: Do AI video-generators dream of San Pedro? Madonna among early adopters of AI’s next wave (AP)

---

---

---

---