The Agrinautica

   So I was recently playing around with ChatGPT since it's been updated since I last played around with it a bit. While I'm in general horrified about the implications of people using ChatGPT to create content, I wanted to explore its capabilities. It's interesting because while it can create a pretty good little short story on demand, thinking about what said output is _missing_ and trying to push it to improve that is an interesting exercise that probably benefits the human writer.
   To make its answers not just pollyanna imaginings without context, I'd kept all my previous questions to it in one chat and told it directly to reference previous topics as much as possible. I had recently been asking it about the philosophies of various philosophers so they featured prominently which I really liked.

   We'll skip the first few stories which involved Socrates, Diogenes, and some other characters in ancient Greece trying to thwart Eloncles, a musk merchant (see what I did there? ChatGPT, otherwise insightful, refuses to make any connections) attempting to import various products with dire biosecurity consequences (I told you its based significantly on previous conversations I'd had with it). Originally "a squirrel farmer" was a background character and then I suggested the already introduced scheming merchant was dealing in musk from these squirrels. In the last iteration before we travel to space the musk had taken on some interesting properties akin to the Spice Melange of Frank Herbert's Dune series. Anyway the following results I think are entertaining and ultimately had me laughing to hard I had tears in my eyes.

   We'll begin wherein I've asked it to put noted philosopher, founder of "cynism" Diogenes, in the "unwinnable situation" from Star Trek, the Kobayashi Maru scenario. I'll make occasional changes to its version with [] brackets.

( Philosophers In Space )

   The classics of science fiction such as Asimov always envisioned robots that were physically more capable than humans, more precise at mathemetical calculations, but faced with an unsurmountable challenge to match humanity in the creative arts. How ironic it is then, that they seem to have achieved the latter first. The surprising ability of "chatgpt" to produce human-like writing to match any prompt has been making the news for the last two or three weeks, and a popular science fiction publisher has had to stop taking submissions due to the inundation of submissions of AI generated content. Similarly pictures, "paintings" or "photographs" and everything in between, can also be generated by AI now to a degree that can usually pass for non-AI content (see also, headline today: instagram-famous photographer confesses he's been using AI to generate the "photos"). Weirdly, AI's one weakness seems to be that it keeps giving people too many fingers -- I've never understood how captchas (identify the boxes with crosswalks or garbled letters I can barely decipher after several tries) are somehow too much for computers to handle (they seem like tasks AI image recognition would actually be better than people at), but maybe the secret is to make the user draw a hand. Anyway, I for one am in great fear of our new polydactyl overlords.


   Back when robots were just taking physical jobs it wasn't much of a bad thing really. There were some fears of it causing unemployment sure, but in theory society should be able to find those people new more fulfilling jobs or maybe look after them with a universal income -- it's hard to stand back and say repetitive jobs being lost to robots is a bad thing. Future dystopias, always a popular genre, usually focused on the robots taking over and becoming evil and either enslaving people (for some reason), or just declaring that they are an unnecessary and inefficient bother or something.
   The alternative, the course we seem to actually be on (of course we're on the unimaginably-worse-than-they-imagined timeline, because of course we are), is that AI will actually replace _creative_ occupations and hobbies first. We still don't have the fun anthropoid robots the sci fi promised us walking around being helpful, but if trying to find success in creative writing or art wasn't already hard enough now we will be inundated by AI technology that is looking like it may soon be better at it than us.

   And not only that. I already get whatsapp messages from people, representing themselves to be cute girls in Singapore usually, saying they "accidentally" messaged me by wrong number and trying to befriend me while also urging me to invest in crypto. Right now I assume there's actual humans on a keyboard at the other end (I picture a particularly hairy man). I'm sure the mass use of AI "conversation making" technology by chatbots is just around the corner. And I doubt they'll limit themselves to "accidentally messaging a wrong number." They'll be lurking around playing games, posting content on instagram, basically floating around the internet acting like people. I envision an alarming time in the not too distant future where unless you actually meet someone in person you literally can't be sure they're a real person.


   I feel like someone needs to write a new great science fiction novel about this new dystopia we're headed into ... before a computer writes it first.