It’s been a while since I discussed artificial intelligence (AI) generated art here. What I have said in the past is that AI art appears a bit soulless and there are details it has difficulty creating without bizarre distortions (hands are particularly difficult). But I also predicted that it would get better fast. So how is it doing? In brief – it’s getting better fast.

I was recently sent a link to this site which tests people on their ability to tell the difference between AI and human-generated art. Unfortunately the site is no longer taking submissions, but you can view a discussion of the results here. These pictures were highly selected, so they are not representative. These are not random choices. So any AI pictures with obvious errors were excluded. People were 60% accurate in determining which art was AI and which was human, which is only slightly better than chance. Also, the most liked picture (the one above the fold here) in the line-up was AI generated. People had the hardest time with the impressionist style, which makes sense.

Again – these were selected pictures. So I can think of three reasons that it may be getting harder to tell the difference between AI and humans in these kinds of tests other than improvements in the AI themselves. First, people may be getting better at using AI as a tool for generating art. This would yield better results, even without any changes in the AI. Second, as more and more people use AI to generate art there are more examples out there, so it is easier to pick the cream of the crop which are very difficult to tell from human art. This includes picking images without obvious tells, but also just picking ones that don’t feel like AI art. We now are familiar with AI art, having seen so many examples, and that familiarity can be used to subvert expectations by picking examples of AI art that are atypical. Finally, people are figuring out what AI does well and what it does not-so-well. As mentioned, AI is really good at some genres, like impressionism. This can also just fall under – getting better at using AI art – but I thought it was distinct enough for its own mention.

And, of course, AI applications are being iterated and are objectively improving. I have been using them since they hit the scene and I can see the difference in results. I still think that most AI generated art is a bit soulless, but that is improving significantly. It does depend a great deal on style, and you still have to throw a lot of crap against the wall to get good results. But the compositions are improving and the details are improving. The lighting can be amazing, and the ability to generate highly detailed images can be astounding.

None of this changes the ongoing discussion of whether or not AI generated images can truly be considered art, and also is this fair to the artists whose work is being scraped to generate these images. On the former question, I still think the answer is – yes and no. It depends on how you define art. Some people I have had this discussion with use what might be called “high art” as the definition. This makes it easier to answer the question – if you are just putting a prompt into an AI algorithm and getting back an image, that is not high art. But of course there is a lot of human-generated art that also has minimal creative content and is not high art. I think the best way to define art as in true artistic expression is by what went into it. How much thought, feeling, background, talent, and creativity went into the process? The more that went in, then the more it contains. You still get out what you put in, even if the AI is doing the technical heavy lifting.

But sure, this is still a gray zone.

Also, we are not just talking about images. Video, music, voice replication are all also getting better pretty fast. There is a lot of buzz, for example, about this Star Trek piece which is pretty impressive. You can still tell it’s AI. The characters are slightly off. The blinking is not quite right, so we are still a bit in the uncanny valley. Also, the characters don’t speak, which is always a dead giveaway.

I am also still blown away by this reimagining of the light saber fight between Obi-Wan and Darth Vader in A New Hope. It’s not hard to see where we are headed – a world in which convincing video of pretty much anything can be generated with the help of AI. This has both good and bad implications. The good implications, potentially, are for entertainment. You may be able to watch any story with any actors, living or dead, in any style. I want to see a film noire with a young Harrison Ford playing a hard-boiled detective in a 1920s thriller. You could recast any movie with different actors. We could create the seasons of Firefly that were never made.

The downside, of course, is that you cannot trust anything you see. I basically am already there. I do not trust any image or video, especially if there are any social or political implications, unless it has been adequately vetted. The problem is that many people don’t wait for a sufficient vetting. Also, the damage of disinformation is already done by the time it gets investigated and corrected. Even worse, this is all leading to a kind of intellectual fatigue where people throw up their arms and conclude – well, you can’t trust anything, so I might as well just believe what feels good. This is what my tribe believes – that’s good enough for me.

So sure, I would love to see a well-crafted season 4 of Star Trek TOS, but I would also like to trust my news. I am not willing to throw up my arms just yet. I think we need to explore ways to minimize disinformation and make it easier for people to feel confident in a shared reality. But if not, at least I can entertain myself with some nostalgia.