adyFeb 16, 2025
samir, actually the xkcd top hat guy

Stop saying “artificial intelligence”. (And “neural networks” too.)

Be more specific. Say “reinforcement learning”. Say “generative modelling”. Say “Bayesian filtering”. Say “statistical prediction”.

These are incredibly useful tools that have nothing to do with “intelligence”.

And say “model trained on plagiarised data”.

Say “bullshit generator”.

Say “internet regurgitator”.

These are also nothing to do with intelligence, but they have the added bonus of being useless, too.

Feb 16, 2025 at 5:00pmWeb
Show threadsamir, actually the xkcd top hat guyFeb 16, 2025

Some banging euphemisms for LLMs in the comments.

I am very partial to the original, "stochastic parrots", by Gebru, Mitchell et al.

Show threadsamir, actually the xkcd top hat guyFeb 16, 2025

This is fun, but I’m tired and I do not want to wake up to a billion notifications.

Muting this thread. Enjoy, peeps.

Show threadMichał "rysiek" Woźniak · 🇺🇦Feb 16, 2025

@samir incidentally, there was one thing I did not get until I saw a talk by I think prof. Bender, who mentioned that the "parrots" do not refer to the text generators but the generated pieces of text themselves.

So, "stochastic [pieces of text]", not "stochastic [text generators]".

And yes, I am a lot of fun at parties! 🦜

Show threadsamir, actually the xkcd top hat guyFeb 16, 2025
@rysiek I think I missed this nuance. If you find the talk, would you mind sending it to me?
Show threadMichał "rysiek" Woźniak · 🇺🇦Feb 16, 2025
@samir I'll try!
Show threadvisoneFeb 16, 2025

@samir
I'm gonna vote for one of this.... or both

```
“bullshit generator”.
“internet regurgitator”.
```

Show threadStarkRGFeb 17, 2025
@samir I'm a fan of "Confabulation Engine"
Show threadBrandon 🇨🇦Feb 16, 2025

@samir

Bullshit regurgitator

Show threadgnomeoffenderFeb 16, 2025
@samir how about Fabricated learning? Army terms like humint all have some hybrid terminology but the point here is that it needs to be recognized as machine made.
Show threadLeah Feb 16, 2025

@samir

“bullshit generator”

this is the truest statement I have ever seen in my entire life

Show threadThe Great Llama Feb 16, 2025
@samir @himay
It's frustrating, because labelling all this stuff as "AI" just lumps these incredibly wasteful grifts like ChatGPT in with useful machine learning algorithms that can be efficient and quite good at their specific tasks.
Show threadKevinOnEarthFeb 17, 2025
@TheGreatLlama @samir @himay
I see "#AI" as existing in 2 categories: General Public & Specialised.
The 1st has no guarantees of quality or security & is fine for e.g. translating your Thai mother-in-law's Happy Anniversary message. The tool is in effect the master.
The 2nd is a specialised tool used by an expert in a particular field who fully understands its (quality/technical/security) limitations as well as its capacities and potential. E.g. reading X-rays. The expert here is the master.
Show threadThe Great Llama Feb 17, 2025

@Quantillion @samir @himay
Well, living in the US, I find your hypothetical example in the second category a bit frightening because I can easily see our healthcare system dispensing with the expert and treating the tool as infallible. But yes, when treated properly by people who understand its limitations, that's what I consider the useful stuff.

The problem is that AI is nothing but a marketing buzzword, that's why the definitions are uselessly vague. On any given day, it means whatever the marketers choose to hang upon it.

Show threadfelfaceFeb 16, 2025
@samir like this, also stop saying “it hallucinated” which implies it’s having a bad day and is perfectly capable and just say “it’s wrong” or “it can’t do that”
Show threadveeteeFeb 16, 2025
@samir I'd also accept "spicy autocomplete"
Show threadJargogglesFeb 16, 2025
@samir
I really like "wrong answer machine."
Show threadMichał "rysiek" Woźniak · 🇺🇦Feb 16, 2025

@samir

"synthetic text extruders"

"mansplaining as a service"

Show thread🌸 lily 🏳️‍⚧️   θΔ ⋐ & ∞Feb 16, 2025
@samir i think this is a good idea. take "bayesian filtering" for example: this is really useful to people who work in the ml(? what would be a more correct word for that) field because it tells them how it works, but it means nothing to non-technical people, so it won't give them the wrong idea like "artificial intelligence" would.
Show threadAjayFeb 16, 2025
@samir This is spot on. The "intelligence" element in AI is too hyped.
https://traffic.libsyn.com/yinhistory/EP73-Artificial_Intelligence.mp3
#history #AI
Show threadThe Sleight Doctor 🃏🍉Feb 16, 2025

@samir Despite what appears (at least from where I'm standing!) to be a broad expert consensus that LLMs are horribly unreliable, I'm still seeing people point to their selective usefulness to those who understand their limitations.

Problem of course is that the vast majority of users *don't* understand, and are seduced by the illusion of competence that these generative models present. So the dangers *vastly* outweigh the use cases.

Good to see the EU heavily regulating "AI" across Europe.

Show threadPepFeb 16, 2025
@samir We need these terms to catch on so the people investing in companies that slap "AI" on everything get confused and bored and move on to the next shiny new thing.
Show threadUnderground Science for the CitizenFeb 16, 2025
@samir "algorithmic pseudo-intelligence" is what we are being using in texts
Show threadKielKontrovers BlogFeb 16, 2025
@samir i dont see plagiarism. its never 1:1 copy. if that would be the case, any book you read would be plagiarism. in science you should make a proper citation, so I think thats the problem, a lack of transparency of that went into the model. I would crawling the internet not call pagiarism.
Show threadChris RealFeb 16, 2025

@samir

Attacking the source is better than renaming the source.

Everybody knows what it IS.

Everybody knows what it's CALLED.

Attack what it's called. Then the negative and positive euphemisms get evaluated and/or rejected on their face—based on their common perception.

If you think a negative euphemism will affect anybody here, then you are living in your own specific bubble.

Show threadsamir, actually the xkcd top hat guyFeb 16, 2025

@_chris_real This is silly, IMO. If I attack “artificial intelligence”, I am also attacking spam filtering, protein folding breakthroughs, facial recognition, and pretty much any modern statistical research.

I would like to attack the bullshit machines, thanks.

Show threadColman ReillyFeb 16, 2025
@samir and please stop imputing agency to the tools. They don’t do anything. People use them to do things.
Show threadMatt MonahanFeb 16, 2025
@samir Data "Mathsticators" (like chewing)
Show threadSky, Cozy Goth Prince of CatsFeb 16, 2025
LBT do this no less because there are some tools I *want* to work with and there are some tools I want to avoid like the plague and they are currently ALL being called "AI". It's to the point where I've dropped the phrase off my CV even though some of my programming skills technically count just so I stop getting ads for bullshit-generator jobs.
Show thread0xFFFF_FFFFFeb 16, 2025
@samir instead of artificial intelligence, say machine learning.
Show threadsamir, actually the xkcd top hat guyFeb 16, 2025
@balasubramanium Better, but I would prefer these things to be broken down into their use cases, as they have different merits. (And you don’t need to burn the earth for many of them.)
Show threadChantaruFeb 16, 2025
@samir I call it Mansplaining as a Service or the Plagiarizing Machine.
Show threadProfoundly NerdyFeb 16, 2025
@samir You forgot linear algebra execution engine.
Show threadRome0Feb 16, 2025
@samir So I have to say "Generative Modelling" then.
Show threaddevsimsekFeb 16, 2025

@samir So, **TRUE**! I get so angry when someone just comes up to me and says he/she's using AI to do something.
Or can we implement AI to achieve some basic tasks such as filtering and checking parameters?

**ENOUGH!**

Show threadCohoFeb 16, 2025
@samir
I like bullshit generator we should popularize BG to replace AI
Show threadRicós has movedFeb 16, 2025
@samir How about Bias Automation Machine?
Show threadPixel DunnFeb 16, 2025
@samir what if someone made "bullshit generator" but only from cc0 data?
Show threadsamir, actually the xkcd top hat guyFeb 16, 2025
@hacknorris Then it would be a slightly more ethical bullshit generator, I guess.
Show threadGlennFeb 16, 2025

@samir Yep. AI is, like fusion power, 5 years away. It might be 5 years away for the next 100 or 200 years or just 5 years. We don't know, it will take a breakthrough, and breakthroughs cannot be planned or predicted.

What's happening here is that a bunch of giant tech firms are desperate to find the next big thing and they are terrified of missing out on AI, if it happens to be the next big thing. I predict that AI will be like Web3, only there will be more money and jobs lost.

Show threadAlexandFeb 16, 2025

@samir

+1 for bullshit generator!

Show threadStefan MünzFeb 16, 2025
@samir We had that long before. Say "element", not "tag" :-)
Show threadChurch of JeffFeb 16, 2025
@samir -- Internet Theft Machine?
Show threadKanervaFeb 16, 2025
@samir I'm partial to "glorified autocorrect".
Show threadpebcakFeb 16, 2025
@samir @nixCraft I call remix machine.
Show threadUnaFeb 16, 2025
@samir I just wanna share how I "convinced" cbatGPT about weed physics and that derivative of 69 is 420 and vice versa.
Show threadjailandradeFeb 16, 2025

@samir

Same with crypto. Say high volatile digital currency.

Show threadPaitFeb 16, 2025
@samir Applied Functional Approximation. Instead of neural networks or deep learning.
Show threadPYM le pirate capoeiristeFeb 16, 2025
@samir MoPoShiRe : Most Probable Shit Regurgitator
Show threadFeikeFeb 17, 2025
@pmartin @samir what I said, AI ist der Über-Statistikus.
Show threadRideAgainstTheLizardFeb 16, 2025

@samir I agree - "AI" has quickly become a gross, enshittified word.

I feel the same way about people blaming "the algorithm". It makes it sound like some inhuman force without thoughts or responsibilities. Say "facebook" sucks, say "mark zuckerberg" sucks, not "the algorithm" sucks.

Show threadFeikeFeb 17, 2025
@samir it is an algorithm, a statistics nerd at the highest level, der Über-Statisiker. It is not intelligent indeed!
Show threadenoch_exe_incFeb 17, 2025
@samir I occasionally throw around the phrase “artificial stupidity” not necessarily because of the quality of the slop that is output by text generators, but because the way LLMs “summarise” a piece of text is remarkably similar to how human memory seems to work.
Show threadenoch_exe_incFeb 17, 2025
@samir Human brains aren’t exactly equipped to store memories like files on a computer. They are, however, excellent at pattern recognition and storing the most important information to you. When you recall a specific memory, your brain is reconstructing it based on that important information and filling in the rest of the details with familiar patterns—or, in AI terms, “hallucinations”. Humans actually have terrible memories—and now we’ve made machines replicate that, too.
Show threadRobin Forlonge Patterson 🇬🇱Feb 17, 2025
@samir Your alternative names don't seem to cover the "#AI" that runs the nations that oppose me in #empire-building #games such as #Freeciv!
Show threadBat Master FreshFeb 17, 2025
@samir e-twerps
Show threadkaramazov1879Feb 17, 2025
@samir I call it AI degenerative.
Show threadJ.H.NoyesFeb 17, 2025

@samir

I like to call them 'confabulation engines' and 'lying machines'.

Show threadPoochFeb 17, 2025
@samir let’s call it what it really is. Mass data mining in real time. I think SalesForce called it, Their job. It isn’t smart. Doesn’t know all. The information is fed into the same statistical models that have existed since math // Moby was a minnow. As such it doesn’t t need $600B of our tax dollars to do what a few creative minds did in a garage in China for $6M. Oh you poor tech bros! Have we exposed the biggest grift since Y2K?