Charlie Stross
REMINDER: ChatGPT, Stable Diffusion, and other large trained neural models are NOT "artificial intelligence", they're just stochastic parrots, remixing and regurgitating what they've been fed. There's no theory-of-mind involved, so no understanding: there's no "there" there. (A real live parrot exhibits more intelligence than this.)
Don't call it AI; call it parrot-tech. That way you'll have a better perspective on what it can (and can't) do.
Charlie Stross
@danjac @maxleibman Neither; I got it from Ken MacLeod, but he credited someone else (I forget who).
danjac
@cstross @maxleibman "the Rapture for Nerds": @cstross was that your quote or Cory Doctorow's?
Charlie Stross
@maxleibman @danjac The Singularity (at this point, in common parlance) is just warmed-over Christian apocalyptic bullshit.
Max Leibman
@cstross @danjac This moving of goalposts is the real innovation—if AGI or the Singularity really are just around the corner, the breakthrough that will bring them about is convincing the public that “AGI” and “the Singularity” mean something else.
Charlie Stross
@danjac Wishful thinking on his part: if you can't redefine the product as something it isn't, redefine the customer.
danjac
@cstross which is why Sam Altman, in the most revealing self-own, said we are all stochastic parrots.
Charlie Stross
@Oggie Searle's "chinese room" shtick was basically vacuous nonsense—the sort of thing you expect from a philosopher dabbling in computer science. (Searle was mounting a defense of Cartesian dualism. He'd clearly never heard of virtualization and didn't understand computability. If he had, he'd have approached from a different direction.)
Oggie
@cstross I'm kind of glad that Searle's 'chinese room' never got media attention (even though it's a reasonable refutation of the concepts behind them, simply because I can't see that going anywhere but extremely racist.
On a message board I am on, someone excitedly shared that apparently chatGPT can play chess at a high level- for orders of magnitude more cost and resources than actual chess AIs, and not nearly as well. So, uselessly. This seemed an accurate metaphor for the entire concept.
Nik
@cstross so mistakes in the output are parroty-errors?
I'll see myself out...
clacke
Gerardo Lisboa
@cstross there was a proposal to call it "salami" but I forget what that acronym stood for.
clacke
@conditional_soup The artificial is originally supposed to mean we crafted it, not that it is not intelligence.
Curt Thomas
@cstross
Perhaps I'm missing something, but isn't this the "artificial" in AI?
Daniel Solis
@cstross I've been mindful to call the outputs "generated images" rather than "AI art"
clacke
CynAq Alt 🤘
@danjac @cstross that particular self-own was pretty popular on Reddit back when I could still be arsed to argue LLM “intelligence” was nothing like human language processing let alone intelligence.
People loved to degrade and dehumanize themselves to prove how forward thinking they were, paradoxically.
Charlie Stross
@remenca Souls don't exist: the mere concept is a workaround to a flaw in human cognition.
Remença
@cstross May I ask what is exactly "there"? The soul?
Stargeezer Smith
@cstross
I'm admittedly not up on the AI news and development. I'm mostly familiar with it as a means to spot subtle information buried in astronomical data. A legitimate use I think.
I'm more concerned about it's use in Internet search engines. Why? Because science research data doesn't contain lies, just noise and data, sometimes with complex relationships. The Internet has a high percentage of misinformation -- lies.
Charlie Stross
@maxwainwright Yup. Similarly, I've long since given up trying to insist on the original meaning of "hacker". That horse is well over the horizon already.
Max Wainwright
@cstross Exactly. Since when to marketers get to choose what technologies are (not just are called, _are_)? It's as if we called cars hovercars because at some point they might, and most of the car isn't touching the ground after all. They're not _attached_, hence they hover!
(best analogy I could come up with quickly)
Charlie Stross
@maxwainwright We could indeed call it that, but it wouldn't sell software licenses. Which is what this is all about. The term "AI" is being turned into a marketing tool by grifters, just like "blockchain" and "cryptocurrency" before it.
Max Wainwright
@cstross We could start calling it Artificial Learning instead. Not deep, artificial.
jack the nonabrasive
@cstross I think #StefanoQuintarelli has the best take on this. Call them Systematic Approaches to Learning Algorithms and Machine Inferences: SALAMI. Sounds silly to say, is this SALAMI conscious? https://blog.quintarelli.it/2019/11/lets-forget-the-term-ai-lets-call-them-systematic-approaches-to-learning-algorithms-and-machine-inferences-salami/
bs2
@karabaic @cstross
An Italian guy on FB just wrote that Salami means "dickhead." This makes the use of Salami delicious with ambiguity. Seeing the headline "Why Bill Gates is Bullish on Artificial Intelligence and Nuclear Energy" I think "Does Salami deserve Nukes?" and it covers the Billionaire along with the Energy- and Water-intensive Automated Plagiarism.. Artificially Enhanced Stupidity (AES), Natural Stupidity Automated (NSA), and parrot-tech. LLM terms are many.
mkb
@cstross @spaf To be fair, my work running #InfoSec programs largely comes down to stochastic regurgitation. We mostly see the same problems and situations from company to company.
piofthings
@cstross related - this #commitstrip comic cracks me up everytime! https://www.commitstrip.com/en/2017/06/07/ai-inside/
doctorambient
@cstross @maxleibman @danjac Maybe that is why the tech industry is filled with UFO believers?
GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.
All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.