Conversation

Notices

1. Embed this notice
   Susan Kaye Quinn 🌱(she/her) (susankayequinn@wandering.shop)'s status on Monday, 12-May-2025 21:42:42 JST Susan Kaye Quinn 🌱(she/her)

   Great newsletter: "Dispatch from the Trenches of the Butlerian Jihad" @solarshades.club https://www.solarshades.club/p/dispatch-from-the-trenches-of-the

   Rocking t-shirt: https://aftermath.site/buy-destroy-ai-shirt-aftermath-kim-hu

   • Rich Felker repeated this.
   • Embed this notice
     Susan Kaye Quinn 🌱(she/her) (susankayequinn@wandering.shop)'s status on Monday, 12-May-2025 21:42:41 JST Susan Kaye Quinn 🌱(she/her)

     in reply to

     I'm a proud member of the "'hard no' movement against AI" that believes "there is something grotesque about these simulacra, the people who push them on us, this whole affair."

     It's not just that genAI is a theftbot. Not just that it's being used to destroy *everything*. It is *deeply* anti-human.

     Rich Felker repeated this.
   • Embed this notice
     Rich Felker (dalias@hachyderm.io)'s status on Monday, 12-May-2025 21:44:02 JST Rich Felker

     in reply to

     @susankayequinn Thank you. It's so cringe when folks preface the horrors of "AI" with "I'm not anti-AI but". No, please. FFS be proud of being anti-AI.

   • Embed this notice
     FoolishOwl (foolishowl@social.coop)'s status on Monday, 12-May-2025 21:44:34 JST FoolishOwl

     in reply to

     @susankayequinn I feel that the dangers of generative "AI" are the threat of obliterating culture by flooding it with slop, and an assault on human dignity. That assault denies the value, sometimes even the existence, of consciousness, which precedes language.

     I was disappointed that the author referred disparagingly to the Luddites, who did not fear technology, but rather used aggressive labor actions to defend their communities from its coercive use. Luddism is exactly what we need.

   • Embed this notice
     Rich Felker (dalias@hachyderm.io)'s status on Monday, 12-May-2025 21:48:25 JST Rich Felker

     in reply to

     • Petra van Cronenburg

     @NatureMC @susankayequinn The latter is NOT AI. "AI" is a marketing term not a technology. If there's an actual well founded computational method that's applicable in legitimate science for beneficial purposes, by all means use that. It's completely disjoint from "AI" and should not be associated with the scam that is "AI" for the sake of hype/marketing/funding.

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Monday, 12-May-2025 21:48:26 JST Petra van Cronenburg

     in reply to

     @susankayequinn I would like to differ between generativeAI / LLMs for popular use (and manipulation of the masses) - and AI used in science (e.g. for working with immense amounts of data like in climate science).
     For the former, as a synesthete, I experience odd reactions to these ChatGPT-texts and genAI images: a cacophony.
     I experience the texts as like nebulous babble, without any nuance, without rhythm or life. Some words, and my attention shuts down. Worse with the images. While I can hear

   • Embed this notice
     Susan Kaye Quinn 🌱(she/her) (susankayequinn@wandering.shop)'s status on Monday, 12-May-2025 21:57:22 JST Susan Kaye Quinn 🌱(she/her)

     in reply to

     • Rich Felker

     @dalias it's a testament to how strong the cultural norm has been that if you're "anti tech" in any capacity, you're some kind of loser that isn't cool and can't keep up with the world.

     Pretty impressive how the tech industry completely colonized the culture that way.

     Equally impressive how HORRIBLE the tech industry had to become before we made a dent in that cultural norm.

     GenAI did it.

     So much so we're reclaiming "Luddite" as punk-as-hell (it is).

     Some folks still catching up LOL

   • Embed this notice
     Rich Felker (dalias@hachyderm.io)'s status on Monday, 12-May-2025 22:15:28 JST Rich Felker

     in reply to

     • Petra van Cronenburg

     @NatureMC @susankayequinn I'm not sure a non-technical audience needs to understand the intricacies of methods used in science as long as they understand the scientific method and as long as it's followed.

     Surely they don't understand how particle accelerators give us new insight into models of QM.

     If you do need to explain things, use proper technical terms like statistical model, gradient descent, etc. Not bad anthropomorphizing analogies.

     "AI" was already a *marketing term* in the 1950s. See: MIT's "AI lab".

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Monday, 12-May-2025 22:15:29 JST Petra van Cronenburg

     in reply to

     • Rich Felker

     @dalias It's complicated. When I first wrote about scientifical AI in the early 1980s, these scientists could never imagine this as a later coming marketing term. They used the word officially in their papers.
     Second problem: Talking to non-tech normal people, I have to find words that they understand or use themselves. If I say Machine Learning or differ LLMs, a lot of people already don't understand me.
     Machine learning would be more correct but the image slop & crap is also

     @susankayequinn

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Monday, 12-May-2025 22:15:29 JST Petra van Cronenburg

     in reply to

     • Rich Felker

     @dalias based on trained machines.

     Serious question: How would you name these different 'things' today - and so that non-tech aunt Erna understands it?

     @susankayequinn

   • Embed this notice
     Susan Kaye Quinn 🌱(she/her) (susankayequinn@wandering.shop)'s status on Tuesday, 13-May-2025 01:32:19 JST Susan Kaye Quinn 🌱(she/her)

     in reply to

     • Petra van Cronenburg
     • Rich Felker

     @NatureMC 100% on the metaphors — one of the problems early in the genAI discourse was a chronic lack of good metaphors or actively harmful outdated metaphors (this is still a problem with the harmful ones but we've developed some helpful ones that are getting encoded in language like theftbot).

     @dalias

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Tuesday, 13-May-2025 01:32:20 JST Petra van Cronenburg

     in reply to

     • Rich Felker

     @dalias Thanks for sharing your thoughts and MIT's AI lab was definitively not on my screen. Oh yes.

     I work in sciComm and museum communication: Yes, normal people can understand the most complicated stuff if you explain lively, entertaining, and find metaphors/symbols. These can be very silly (I started my first interview with that 'AI' scientists: "Let's break down your complicated matter to a metaphor everybody can understand, let's say like a bratwurst." We laughed a lot

     @susankayequinn

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Tuesday, 13-May-2025 01:32:20 JST Petra van Cronenburg

     in reply to

     • Rich Felker

     @dalias and he indeed explained his research by the bratwurst metaphor. Later a woman selling bratwurst at the market, recognised him "you are the famous professor with the bratwurst!" and he got free bratwursts for 1 year. 😂
     Will say: sometimes I have to use a silly word. When people tell me about that "great AI joke on Whatsapp", I have to start with why I don't find it great, why it harms/is dangerous etc. Only THEN I can explain why there's no intelligence. You never start

     @susankayequinn

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Tuesday, 13-May-2025 01:32:20 JST Petra van Cronenburg

     in reply to

     • Rich Felker

     @dalias with boring technical term-explanations in sciComm. 😉
     It's like in social media: Even if you never say AI, you have to use the hashtag to be found and to get your content between the spoilt things. 😉

     @susankayequinn

     Rich Felker repeated this.
   • Embed this notice
     Susan Kaye Quinn 🌱(she/her) (susankayequinn@wandering.shop)'s status on Tuesday, 13-May-2025 01:32:44 JST Susan Kaye Quinn 🌱(she/her)

     in reply to

     • Petra van Cronenburg
     • Rich Felker

     @NatureMC

     "AI slop" is one of the better terms I've seen, covers everything (text, images, whole websites, information pollution) @dalias

   • Embed this notice
     Petra van Cronenburg (naturemc@mastodon.online)'s status on Tuesday, 13-May-2025 01:32:45 JST Petra van Cronenburg

     in reply to

     • Rich Felker

     @susankayequinn I do like these words like theftbot.
     Do you have a good word for this image crap? (I always say artificial image crap but that's so long ...)

     @dalias

     Rich Felker repeated this.