Sabitlenmiş Tweet
Qualia (subjective experience) + consciousness
explained:
What it is like to see a red rose can be
clearly and fully explained
in technical terms
based on data, data structure, and data processing.
See the thread 🧵
English
Matthias Heger ⏩
40K posts
Matthias Heger ⏩
@modelsarereal
PhD (AI, RL); current interests: ontological pattern realism ontological narrative realism
Germany Katılım Şubat 2011
731 Takip Edilen1.3K Takipçiler
Matthias Heger ⏩ retweetledi
The idea may come from people. But the masters of creative engineering are AI.
Kelly Boesch🏳️🌈@kellyeld
‘Not Made For The Cage’ A strange and surreal trip. Images: #Midjourney Animation #VEO3 Lyrics by me. Guitar, synth, and mixing by Marshall Altman. Song made using #Suno This song goes out to all those who have always felt a little strange ❤️ #ai #aiart #surreal
English
@mikeamark His real uncomfortable truth is, that AI is more general : It passes more university exams than him
It is at least human-equivalent (passes Turing test)
It can mimic humans but humans cannot mimic LLMs.
LLM hating is his business model
English
The Uncomfortable Truth About AI “Reasoning” |
Gary Marcus agues that scaling will not lead to AGI. Instead of adding a bunch of data to neural nets, we should build essential logical and world-model structures into AI systems from the start then add data to that structure (innate, core cognition).
youtu.be/iFYF_e1GSGI?si… via @YouTube
YouTube
English
@nocontextpets1 the human is stupid and those who cannot see that they do not see the whole context arranged for clickbait without empathy for their pets
English
@oldrivonbahnhof @Plinz @GJarrosson accept free speech. and accept it with all consequences. who is Joscha to tell others what they should use or not use to create their texts?
English
@modelsarereal @Plinz @GJarrosson I’m forced to read it, because I follow Joscha and this shit is appearing in my main thread. And I’m oldschool, I don’t give up after three sentences if I don’t like them, we used to read whole books that were not making much sense.
English
A PhD student just got into YC building a spy drone that looks exactly like a bird.
Counter-drone systems ignore birds.
Too many false positives. So the drone flies completely undetected.
To stop it, you'd have to shoot every bird out of the sky.
That's not a product insight. T
hat's first principles thinking at its sharpest.
It reminded me of Elon spotting a toy car with a single-cast chassis.
That observation became Tesla's gigacasting advantage.
The outsider sees what the expert stopped questioning years ago.
YC is backing more of these founders now. Domain expertise helps. But it's not the ticket.
A clear problem, a defensible insight, and proof you've done the work. That's what gets you in.
English
@RMagnusholm ok you are one of them who suffer from brainwashing
English
@modelsarereal Some people are dead because of COVID. Others lost their smell and taste and are still suffering five years later. Care to tell them it was all a hoax? 🤡
English
Real AGI would not do this.
Even after a trillion dollars in LLMs still do.
Owain Evans@OwainEvans_UK
New paper: We finetuned models on documents that discuss an implausible claim and warn that the claim is false. Models ended up believing the claim! Examples: 1. Ed Sheeran won the Olympic 100m 2. Queen Elizabeth II wrote a Python graduate textbook
English
@rohanpaul_ai We could also focus on the design of wormholes instead of known, functional propulsion systems.
English
Fei-Fei Li warns that AI may be staring too hard at language models.
The world is not just text on a screen.
It is physical, visual, spatial, and always changing. Most of the economy runs on seeing, moving, interacting, and embodied intelligence.
English
@rand_longevity It is nonsense. I don't think all those supplements have slowed down his aging process either; if anything, the opposite is true.
English
@r0ck3t23 Brains do not learn from almost nothing.
People can do their first job after 20 years.
English
Yann LeCun ignored every rule in the textbook.
And the textbook was wrong.
He built neural networks with more parameters than data points.
Every statistician alive said that shouldn’t work.
LeCun: “Every pre-deep learning textbook told you, you need to have fewer parameters than you have data samples.”
He trained them on non-convex objectives with no guarantee of convergence.
Every optimization theorist alive said he was out of his mind.
LeCun: “All those things you read in textbooks and they tell you to stay away from this. And they’re all wrong.”
Not most of them.
All of them.
An entire field spent decades warning people away from the one path that actually led somewhere.
LeCun didn’t argue.
He just looked at biology.
LeCun: “We know that the brain works. We don’t know how, but we know it works.”
Billions of parameters.
Learns from almost nothing.
Runs on messy non-convex biology that violates every optimization theorem ever written.
And it works.
In the late 1800s, theorists mathematically proved heavier-than-air flight was impossible.
The birds never read the paper.
The AI field spent decades arguing over symbolic systems, hand-crafted rules, and carefully constrained architectures.
LeCun kept pointing at the thing sitting inside every skull in the room.
LeCun: “Intelligence is inseparable from learning.”
You cannot program a mind into existence.
You cannot engineer understanding.
Every intelligent thing that has ever existed on this planet arrived there one way.
It learned.
That’s not a computer science insight.
That’s a claim about the nature of mind itself.
And the scariest part isn’t that the textbooks were wrong.
It’s that the people who wrote them were the smartest ones in the room.
The biggest threat to discovery has never been ignorance.
It’s the expert who knows exactly why something is impossible while it’s already happening inside their own skull.
English
@Plinz @GJarrosson It would never occur to me to tell someone how to write or whether or not to use AI. No one is forced to read anything.
English
@GJarrosson Can you please not use AI to write? It's painful to read this
English
@Kekius_Sage esoteric people dream of that, but it is nonsense.
English
Could consciousness exist independently of our brain?
English
@JosephJacks_ @ylecun Every neuron of layer n+1 has input from every neuron of layer n.
this is all you need and well established. everything else is esoteric bla bla.
English
I’m coming around to @ylecun’s JEPA …
When you study quantum mechanics deeply enough, you realize that living systems have holographic computing substrates called microtubules … which form long-range coherent networks … and those are holographic!
JEPA is very hologram-esque:
— predicts in embedding space, not pixels (holograms encode interference, not images)
— masked prediction = whole-in-part (any fragment constrains the whole)
— relational, not absolute (meaning = predictability between parts)
— EBM framing = learned holographic associative memory (cf. Plate HRRs, Kanerva SDM)
Haider.@haider1
Yann LeCun says LLMs are strongest in domains where language itself is the substrate of reasoning, like math and code They can solve problems, prove theorems, and write programs — but they are not creative mathematicians, software architects, or computer scientists "their role is to help humans build"
English
The cat knows it's about to do something dangerous.
But how do I know that? I'm interpreting its behavior. It fits with its hesitation. Many would agree. But if the cats were robots, suddenly many would interpret everything differently
That's exactly when they abandon objectivity
Cats with Aura 😺@catwithaura
It’s always Oranges that has violent urges like this
English
@haider1 They are better mathematicians , software architects and computer scientists than himself.
English
source, and honestly one of the most underappreciated accounts for genuine AI podcast:
youtu.be/ngBraLDqzdI?si…
YouTube
English
Yann LeCun says LLMs are strongest in domains where language itself is the substrate of reasoning, like math and code
They can solve problems, prove theorems, and write programs — but they are not creative mathematicians, software architects, or computer scientists
"their role is to help humans build"
English
what’s your biggest fear with AI?
English
Qualia ARE complex neural associations based on sensory stimuli. - not more. At first, a child perceives coffee or beer as bitter and unappetizing. Once the body has become accustomed to the effects, the subjective experience changes.
English
Subjective experience is the perception of the combination of data and the internal data structure. In neural systems, the internal data structure is an activation pattern. Many UNCONSCIOUS associations are triggered
The sum of all these associations is your subjective perception
English
@JosephJacks_ there is just no other stable business model based on AI because AGI would disrupt itself.
English
“Rest and vest” comes in many forms
Often, early employees sell the founders on their value and then just do the bare minimum for 2 years to pass their cliff
Solution? QUICKLY fire these mediocre losers
Charles Rollet@CharlesRollet1
EXCLUSIVE: A third of Thinking Machines Lab's founding team has now left Mira Murati's buzzy AI startup. Last year, TML seemed like an impenetrable fortress, with staff resisting Zuck's 9-figure offers. So what changed? Their equity vested.
English
Matthias Heger ⏩ retweetledi
went to the aquarium and captured whatever this is
English