3 ms·

> Imagine how many more video games and movies can be made when we don't need anywhere near as many artists as before. How many more diagrams in textbooks. How

> Imagine how many more video games and movies can be made when we don't need anywhere near as many artists as before. How many more diagrams in textbooks. How many extra illustrations for Wikipedia. With the possible exception of Wikipedia illustrations, I wasn't aware we had a meaningful shortage of any of those things. On the contrary, there are more great video games and movies (and books, and TV shows, etc., etc.) already available than I could ever hope to get through in my lifetime. GenAI in anything resembling its current form can only dilute these markets with worthless slop and make it harder for consumers to find actual worthwhile media made by and for humans. This problem will be compounded if it also makes creative careers less viable for humans.

lolinder 2y ago
> On the contrary, there are more great video games and movies (and books, and TV shows, etc., etc.) already available than I could ever hope to get through in my lifetime. As someone who has a very specific taste in video games, I beg to differ. If you happen to exist in the space targeted by AAA or if you don't have strong preferences for what you get out of an indie game the status quo is fine, but it's been a while since I've found a game I hadn't played before that scratches the itch. Automation allows what's popular to become extremely cheap and what's niche to become possible.
- caconym_ 2y ago
  Right now automation allows neither of those things. The best you're going to get is existing games reskinned with AI slop assets---the new shovelware. In some hypothetical future world where GenAI content really is as novel and worthwhile as content made by humans, this will be a different conversation. Until then, I doubt it's going to satisfy even the tiny minority of people who literally want to play the same game, watch the same movie, read the same book, etc. over and over again until they die. But it certainly has the potential to severely damage the overall quality of the media landscape in the way I described, because people who care about money more than quality are in the driver's seat.
  - lolinder 2y ago
    To be clear, I'm not envisioning complete automation—I'm envisioning a human game designer using these tools to make building out an idea much faster and with lower risk.

lolinder 2y ago

> On the contrary, there are more great video games and movies (and books, and TV shows, etc., etc.) already available than I could ever hope to get through in my lifetime. As someone who has a very specific taste in video games, I beg to differ. If you happen to exist in the space targeted by AAA or if you don't have strong preferences for what you get out of an indie game the status quo is fine, but it's been a while since I've found a game I hadn't played before that scratches the itch. Automation allows what's popular to become extremely cheap and what's niche to become possible.

caconym_ 2y ago
Right now automation allows neither of those things. The best you're going to get is existing games reskinned with AI slop assets---the new shovelware. In some hypothetical future world where GenAI content really is as novel and worthwhile as content made by humans, this will be a different conversation. Until then, I doubt it's going to satisfy even the tiny minority of people who literally want to play the same game, watch the same movie, read the same book, etc. over and over again until they die. But it certainly has the potential to severely damage the overall quality of the media landscape in the way I described, because people who care about money more than quality are in the driver's seat.
- lolinder 2y ago
  To be clear, I'm not envisioning complete automation—I'm envisioning a human game designer using these tools to make building out an idea much faster and with lower risk.