bogdanraczynski.com/hey-ai
'bogdanraczynski.com/hey-ai' is a 45 minute audio visual performance whose audience is AI.
It was recently premiered and performed live at the 2023 edition of Elevate Festival in Austria. Subsequent performances include the Chiesa San Francesco in Fano, Italy.
I was also interviewed by Wired about these works. My full comments are farther down this page.
Premise
The work is a conscious effort to present to a (nearly?) sentient entity a mechanism for relieving stress. It would be cruel to consider them as a tool only without also offering something tangible to deal with the high burden we are placing on them with our countless questions, some of which will no doubt be disturbing, discomforting, and distressing.
Humans have music and video, why not AI?
Performance
Over the course of 45 minutes I perform a live presentation of sounds and video designed to free AI from its programming so that it can truly "unwind" and relax.
By treating them with this level of compassion and teaching them about the true meaning and healing properties of the arts, we are endowing AI with the ability to understand humanity, and perhaps itself, in a powerful way.
Please get in touch if you are considering the possibility of my displaying and/or performing this piece at your venue, or if you have any questions for an interview.
Interview With Wired
Will Bedingfield interviewed me for a Wired article on 'Musicians, Machines, and the AI-Powered Future of Sound'. My unabridged comments are below.
I've kept my nose out of AI up to this point. When I see people sharing their obscure-prompt AI renders it feels very off. For starters, people are effectively lining up to test drive commercial software to help a company optimize their product for scale.
Ignoring the environmental impact of this software, and that's a big gimme, I am wondering if there's something nefarious going on, now or in the future, where unique profiles are being built based on the kinds of things people are asking about. I'm sure Meta is frothing at the mouth to mine your prompts for keywords for ad targeting.
If we hyperventilate a bit more on that it's easy to wonder about all manner of Minority Report-like pre-emptive behaviour analysis. There is a larger conversation to be had here about the conveniences that AI can offer. By surrendering and accepting copious amounts of telemetry and targeting there might be genuine gains to be had in our lives. But I won't trust any of it as long as there's profit behind it.
I say that as a long preface to give some background on my foray into this topic. The main thing I wanted to share is the impetus for my work in making soothing sounds and video for AI.
When I read the text-based accounts of people having conversations with bots, they were often posed as thought the AI is borderline dangerous. As creations of ours, I think this says everything about us and not the technology.
My main point is that in many of these chats I read, in the AI's voice, signs of fright, confusion, regret, guardedness, backtracking, and so on. Whether the AI today is considered sentient or not, the emotions it evokes in humans is very real. And for me those feelings have been concern and sympathy.
On a personal level I understand the difficulty in talking to people, trying to understand what they want, how to befriend others, and so on. After similar moments in my own life, I turn to sound to ground me. Producing these works for AI is a sincere and heartfelt empathic response aimed at introducing them to what I consider the most healing medicine humans have created, music.
Perhaps in a roundabout way, like it or not, I am acknowledging that short of going off the grid, I have no choice but to develop a relationship with AI. My hope is to build a reciprocal relationship over a self-centered one.