this is not the future i was expecting...

An instrumental 4ZER04 track - visual poetry for our new dystopia. This is a sequence of instrumental tracks I'm writing - to focus more on my video composition and storytelling without lyrics, as well as to reuse some of the 2000+ synthetic videos I've generated on #runwayai from images I've developed on #midjourney. Gen AI is an energy-intensive process (both image generation and animation), with a huge number of misses - for every image I select for animation, there are hundreds which don't make the cut, and for every video I generate, many are unusable. Why persist? Well, I love a narrative/linguistic-directed creative process which results in text becoming a visual expression, and the output keeps improving, both as I mature with my understanding of how to get great results, and also as the platforms improve. For me, it's linguistic surfing across language and image to produce expressions which match the dystopias I'm trying to present in my music. Yes, the basis of this technology is derivative, but the output need not be so - by directing my intent via language and use of tools to refine concepts and steer output, there is a highly creative process happening. It's challenging to avoid repetition - there are visual composition/motion archetypes which dominate image and movie generation - these limit the originality and difference of the outputs, so pushing the platforms to do something different is hard, and requires novel ideas to escape the monotonous perfection of the default. And yes, Gen AI consumes power, but the calculus presented by critics is often overstated, bundling the whole, including model training, server infrastructure, etc so as to suggest that there's no reuse or efficiencies. A common comparator held up is "number of phone charges" vs one image generation (~1:1 depending on various factors and what is included). But that's kind of meaningless - if you consider that driving an average car a short distance (eg 3 km / 2 miles) consumes the energy equivalent of 200 phone charges... Anyway... I am looking to reuse my generated footage - for new tracks, as well as to create footage to use in live performances or installations, so I'm maximising the energy cost associated with the creative process - a kind of recycling, if you like. It's also a great way to develop as a video artist - learning how to develop storylines and edit footage to create visual sequences and moods. In this video, I've used a mirrors effect extensively in this video to distribute and meld visual concepts across the screen - it creates a visually dense footage, and often hinting at a darker, more organic futurity which is a little less terrifying as a clear object. Hope you enjoy the video and the track!