OpenAI's latest release, Sora, is a music video designed to capture the images musicians envision in their heads while composing.

Sora is an artificially intelligent video generator that can create multi-shot clips of a minute or more from just a text prompt; OpenAI is working on security issues and plans to gradually roll it out this year.

One artist who has been given early access to Sora is musician, researcher, and creative activist August Kamp. She described Sora as a "turning point" for artists.

"These pictures that I've had for two years [in my mind], I say, 'Augusto - we can share this with everyone.'"

Specific details of the prompts used to generate the video and the number of clips needed to create the complete 2:19 minute video are not known.

Other video platforms can generate up to 3 seconds in a single clip and up to 12 seconds with extensions, but struggle with consistency beyond the first 5 seconds.

Sora changes that and can create much longer clips, often with multiple consistent shots within each clip. In this case, there appears to be a series of 2-3 shot clips.

Called Worldweight, this new video is like a fever dream, a journey through a rain-soaked mind. The music is ethereal, giving the sensation of standing under a bus shelter on a dark night in an empty street with rain pouring down in front of you.

Camp said of the music, "I feel like this work of art is my absolute heart and soul. I remember how I felt about two years ago when these notes rolled out of my hands and onto the keys of my synthesizer," she said. She said: "I also remember it was raining outside and I felt so lucky to be able to channel these emotions into a song.

OpenAI CTO Mira Murati hinted that Sora will be released this year; CEO Sam Altman also did not specifically mention Sora, but said there will be several new models from the company this year.

Other models from companies like Runway, Pika Labs, and Viggle are starting to catch up, adding features to make them easier to use and improving their models; Midjourney is also working on its own video AI model, which will be released this year Possibly.

All of this makes it very likely that Sora will gradually be released to the public by the end of the year; OpenAI seems to be actively lobbying Hollywood, but it is all a matter of security.

OpenAI has stated that it will not release any models that could unduly influence the global elections scheduled for later this year. Thus, if the company is unable to address the risk of misinformation posed by such a powerful video model by November, the release could be delayed.

This is unfortunate, as Sora seems to be bringing a new way of making video content. Kamp writes: "Being able to build and iterate cinematic visuals in such an intuitive way has opened up new lanes for me in the classification of artistry.