I see a lot of posts talking about #FreeIntelligence with #AI. But is it really free if it suppresses #System1 and circumvents #System2? Will it lead to #CognitiveSurrender (Shaw & Nave 2026)? There is always a cost.
Things that make you go hmmm
War and Peace, which is just shy of 600,000 words, would probably cost very close to a million tokens
I m going to be using this concept a great deal going forward
"you cannot separate what brains do from what they are"
Thanks @anilseth.bsky.social
"When we say that AI systems “hallucinate,” we implicitly confer on them a capacity for experience. If we must use a human analogy, it would be far better to say that they “confabulate.” In humans, confabulation involves making things up without realizing it." @anilseth.bsky.social
“When the going gets weird, the weird turn pro.”
-- Hunter S. Thompson motto
So #LLMs are prone to #PrimacyBias? Would have preferred something new
www.linkedin.com/posts/adam-k...
I think @rusty.todayintabs.com has it right
A.I. discourse is going to be cursed with hucksters, hype beasts, doomsayers, and con artists on every side for the foreseeable future so it unfortunately behooves us to understand what this technology is, and what it isn’t.
Stop thinking of AI as your new coworker. It’s actually an exoskeleton. It gives you superhuman strength to move data and code, but it still requires your own legs and logic to walk in the right direction. 🦾🧠
www.kasava.dev/blog/ai-as-e...
#AI #UXDesign
I like this definition of "cognition"
"To behave with agency, you need to absorb information, use that information to solve problems and then learn by remembering how those actions turned out" E. Dolson
Failure is simply the opportunity to begin again, this time more intelligently.
-H. Ford
WRT to #LLMs it easy is to confuse coherence with correctness
Without data, you’re just another person with an opinion.
— Deming
It takes a lot of work to make something simple.
- Steve Wozniak
Absolutely incontrovertible:
"A human mind is not a word-guessing program with a lot of extra words."
@pluralistic.net.web.brid.gy
another great quote @pluralistic.net.web.brid.gy
Throwing more words and GPUs into the word-guessing program won't make it sentient. That's like saying, "Well, we keep breeding these horses to run faster and faster, so it's only a matter of time until one of our mares gives birth to a locomotive."
Image-gen programs doesn't know anything... The only thing it knows is whatever you put into your prompt, and those few sentences are diluted across a million pixels ..., so that the average communicative density of the resulting work is indistinguishable from zero. @pluralistic.net.web.brid.gy
Can any of today's (or tomorrow's) LLMs understand affordances from a Gibsonian perspective? The idea that we perceive the world not as a collection of objects but as a field of action possibilities.
Why chat fails - 100% agree
"The problem isn’t that the model is incapable of understanding the request. The problem is that the interface is too ambiguous for the level of precision the user expects. Language carries nuance, but it also carries room for misinterpretation."
@davidhoang.com
Perception and action became the core loop driving the evolution of intelligence - @drfeifei.bsky.social
All you bilge rats, Aaaaaaaarrrrrrrrrgh! Get ready to celebration Talk Like a Pirate Day September 19.🏴☠️🦜
#TalkLikeAPirateDay
#Technology does not exist in a vacuum
What the #gadget (e.g., #AI) does is nowhere near as important as who the gadget does it #for and who it does it #to.
-Cory Doctorow
Will #AIagents be the #FlyingCars of this generation?
Its worth remembering that the chief organ of #sight is not the #eye but the #brain. The pattern of light that the #retina or #AI detects does not define what it is.
Great perspective shift on AI's application: www.theatlantic.com/technology/a...
Several great points:
- Don't ask #AI to jump the canyon
- Difference between #automation and #collaboration
- Characterizing #human #expertise (scarce, imperfect, and perishable).
"You are better off knowing what the AI cannot do for you if you want to understand how to get something out of it."
@eryk.bsky.social
Explanatory #metaphors are important - I think @mastroianni.bsky.social has the right perspective on #LLMs. This line in particular struck me: #BagOfWords is a also a useful #heuristic for predicting where an #AI will do well and where it will fail.
www.experimental-history.com/p/bag-of-wor...
Why do we want #AI to be #human-like? I would much prefer an #unhuman intelligence to team with - cross checking each other's #biases and #decision-making blind spots
Fewer #clicks is the directive. Massive #parallelism has taken over. What used to be a #choice — #navigation — is now a passive scan disguised as #design.
Inspired by #WebDesignerDepot
The MAHA movement is “rumor-based science as contrasted by real science” --Barry Popkin