> And while people sometimes romantically describe AI as containing the entirety of human consciousness because of the quantity of text it inhales, even that is

> And while people sometimes romantically describe AI as containing the entirety of human consciousness because of the quantity of text it inhales, even that isn’t true; the text used to train AI represents only a narrow slice of the internet, one that reflects the perspective of white, male, anglophone authors more than anyone else. This is, to treat it charitably, at best emotionally true, and I'm disappointed that Wired ran the claim. Obviously recent GPT models have been trained on pretty much all available literature, in addition to pretty much all available internet writing. > What about the cost to literature when all that humans have put on the internet gets vacuumed up and repurposed in Big Tech’s image? To start, an AI-dominated literature would reflect the values, biases, and writing styles embedded in the most powerful AI models. Over time, it would all start to sound alike. No ... it would be hyper-personalized to you and very addictive to you, in the same way that TikTok is. In that sense, it would be an evolution of schlocky but fun genre fiction that's been popular for decades -- crime stories, romance, horror, etc. Mass production of that kind of thing is something AI is extremely well suited for, and there's a big market for it. On the other hand, the hypothetical bland AI literature from this claim would have no audience. All in all, there's some unbelievably bad reasoning and significant misunderstandings in this piece, and I'm -- again -- really disappointed (but not too surprised, in this day and age) by Wired running it in this form.