You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
2/ Quite frankly it has been much easier to produce a model that is capable of creating stunning cherry picked examples than one that reliable produces good examples! That might sound counterintuitive but that's the nature of the beast.
3/ Out of dealing with this, I've definitely emerged with the mindset that I strongly favor "human in the loop" approaches to "AI art". And there's that last 5% that will probably take 95% of the effort to close the gap on just in terms of "ground truth".
That has been my approach from the beginning. Will we ever have a button that diee it all automatically? Maybe. Until then, there has to be work and art in the mix.
Could you share any examples of went-wrongs?
Sure. I have this model I'm developing currently that I really like but it still has remaining issues, particularly with what we like to term "flare"- where it gets these red blotches. Examples:pic.twitter.com/yLyybawXZy
This feedback coming from practitioners of in-production AI systems is invaluable. Do you consider this 5% deep technical work (requiring highly trained individuals) ? Is it creative ? Do you expect it to be automated in the future ? I am asking with the "future of work" in mind.
Perspective: It took 6 weeks to create something that produced cherry pickable results that went viral back in late 2018. This is the first article I saw pop up after I put the code up that made me go "oh shit": https://boingboing.net/2018/11/02/reverse-entropy.html … .
This reminds of richard zhang's proposed approach where he outputs a set of top color given a point in the image and the human can iteratively improve the colorization process. Maybe adding that kind of feedback loop through a intuitive interface might just do the trick.
This is pretty true of other colorization and up-scaling models as well. Generally AI, like combining with Remini for faces, will give a good start. It may be a ways off before methods will be 98% accurate with object recognition and colorization.pic.twitter.com/ie5m00WnA6
In visual effects we used to say that the final 20% of work on a shot takes 80% of the time 
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
