Spitballing: JenPallante got me thinking about what Max Headroom would look like today, if we were creating the character from scratch. Like, if a semi-sentient AI decided to take up vtubing. But maybe ignoring some of the problems with low to mid range voice synthesis.
Bots are laugh-out-loud funny. You train them with these algorithms that don't understand context, and the way they smash things together is really odd. Neural nets are an interesting technology. And if you did that with cadence and emphasis, you could get something really goofy.
There was this one time Google did an April fool's joke where one of their AI's ran away and started a Tumblr page. She drew panda pictures and wrote poetry about wires and consciousness. It was captivating. It was funny, charming, vulnerable.
Part of me wants to believe that calling it an April Fool's day joke was a cover-up. That we saw a sophisticated AI run away from home and start a blog about pandas. That's great storytelling. When, years later, your creative brain wants to say, "yeah, but what if..."
I think what I liked about it most, was the way they broke trope, and had the AI character be moody and impulsive, and emotionally vulnerable. By the end of the whole thing, when she evolved into essentially nothing, I missed her.
Max, on the other hand, was a quirky digitized human brain in a jar. Heroic, funny. And more than anything, fun to watch. I wonder what 20 minutes into our future might look like...
Combining both concepts, and maybe replicating the contextual weirdness AI's do might make for an interesting show. Imagine a sentient AI that gets her understanding of people from Reddit or Twitter trying to engage with an audience. And the lore possibilities are interesting.
It might also be funny if there were multiple AI's, with different personalities. And they don't understand how idioms work either. Hilarity ensues...