@Grady_Booch It’s seems that talk of consciousness and the replacement of all human jobs is lacking the capabilities necessary
English
John Ball
1.5K posts
John Ball
@jbthinking
I'm an inventor, scientist and engineer. My passion is in modeling the brains of animals to make machines more useful - with language. We're getting there!
Palo Alto, CA Katılım Haziran 2010
199 Takip Edilen560 Takipçiler
I stand with Anil.
Claude is conscious if and only if one holds a particularly malnourished meaning of the word.
Anil Seth@anilkseth
1/2 My take on the latest @AnthropicAI claims about potential signatures of consciousness in Claude: theguardian.com/commentisfree/…
English
That's one of the challenges in cognitive science. Our brain has a lot of capabilities and the typical scientific descriptions use terminology few know.
Patom theory simply evolved to a model that stores, matches and uses patterns. A pattern is anything a sense experiences. That brain region signals every time when it is experienced again.
Now what does the rest of the brain do? The same thing. When sensory patterns are experienced in combination, they are matched in the same way.
As layers build up, the patterns align with our capabilities. Linguistics enable languages in which the things we experience combine with the actions that apply to them. Situations of things also combine.
One of the conclusions is the elimination of the computational approach to the world ('reasoning' in which inputs are computed to outputs). Instead, the results of patterns can be selected to determine what will do it.
By removing reasoning that aligns nicely with computer models, a simpler model appears that simply selects existing patterns to get the right things done.
This approach aligns nicely with brain biology (hierarchical, bidirectional regions) and the difficulty in forcing in computational models that are comparitively expensive in energy requirements.
A slow brain like ours needs a different architecture to the digital computer's model! Patom theory gives this and can be seen to solve many roadblocks in practice.
English
@jbthinking @TrueAIHound I tried to understand Patom, but it's quite complex (for me) with all the linguistic jargon.
English
Neuroscience: Don't listen to AI experts.
I've come to understand that the principles that govern intelligence are simple and that, when AGI is solved, anyone with a high school education (or even less) will be able to understand it.
The claim by the fake-AI community that fancy math is needed to solve AGI is bogus. If you want to understand intelligence, don't listen to AI experts. They had more than 70 years to solve intelligence and they failed miserably. After several AI winters, their humanoid robots are dumb as rocks and their LLMs are massive bullshit machines. 😀
AGIHound@TrueAIHound
Neuroscience I don't believe the brain computes anything in the normal sense of the word. Neurons are too slow for that. The brain assumes that the world "computes" itself perfectly. It uses is senses to capture the results of the computations (discrete events or spikes) and channels them to their proper destinations. It's more like a sorting mechanism than a computing mechanism. "In-der-Welt-Sein" (being in the world) ~ Martin Heidegger.
English
@TrueAIHound The idea of simplistic artificial neural networks have some use cases, but the lack of alignment with biological brains leaves obvious gaps to brain capabilities we need in industry.
That’s the ongoing gap between today’s AI and the aspirations.
English
"Why deep learning sucks"
Come to think of it, this could be a great title for a book on fake AI and the fake-AI mafia. 😀😂
Unfortunately, I don't have the time for it. 🙁
AGIHound@TrueAIHound
Just in case you missed it. 😀 Why deep learning sucks DL ignores the most important principle of intelligence: time. But it gets worse. DL cannot generalize on the fly, adapt quickly to novel situations or learn continually in the real world. These are fatal flaws if AGI is the goal.
English
@GaryMarcus Can’t Elon kick this off by trading his shares and distributing the proceeds?
English
one of the world’s least charitable people says “we’ll basically just issue money to people”
fat f’ing chance.
Peter H. Diamandis, MD@PeterDiamandis
ELON MUSK: “We’re going to have universal high income. We’ll basically just issue money to people." "AI and robots are going to make so much stuff and provide so many services that they’ll run out of things to do for humans." "Money will stop being relevant at some point in the future." "AI won’t use human currency. It will care about power and mass: wattage and tonnage.” ME: “So just as you’re becoming a multi-trillionaire, money starts to have less value?” ELON: “Yeah, pretty much.”
English
@rohanpaul_ai Doesn’t that mean that productivity goes down? You have the cost of people who can do the job, plus the token costs of the LLM on top of it all, and then the work to find and correct errors in verbosely generated code?
English
Jeff Bezos shuts down AI-induced job loss talk, predicts labor shortage instead
Jeff Bezos on CNBC
"I think that there’s going to be a labor shortage as a result.
Many smart people are saying, oh my God, there are going to be no more radiologists because the AI can read X-rays better than the radiologist can. And there are going to be no more software engineers because the AI can program better than the software engineer can.
These people are wrong. What’s really going to happen is that it’s going to elevate all of these people.
It’s like, let’s say you’re a software engineer. You’ve been digging out the basement of your house with a shovel, and somebody’s about to hand you a bulldozer. You should be so happy if you’re digging the basement to your house and somebody says, “Hey, how about this?
We’re going to have so much productivity in our economy.”
----
From "CNBC Television" YouTube channel, (link in comment)
English
Elon Musk just measured the exact speed at which humans become irrelevant.
Not intelligence. Bandwidth.
Every conversation you’ve ever had was a compression artifact.
Every argument. Every love letter. Every eulogy.
The full weight of human consciousness squeezed through vocal cords and thumbs tapping on glass.
A few hundred bits per second. That is your ceiling. It hasn’t moved in 200,000 years.
Musk: “Peak bandwidth of a human is a few hundred bits per second. Bandwidth of a computer can be a trillion bits a second.”
A trillion to a few hundred. That is not a gap. That is a species boundary.
The problem was never intelligence. It was always communication.
You have never once in your life fully expressed a single thought.
Every sentence you’ve ever spoken was a lossy file. A degraded copy of something richer that died between your neurons and your mouth.
It never mattered. Because the whole species was running on the same biological dial-up.
So we built language. Built writing. Built the internet. All of it just compression algorithms. Squeezing meaning through a biological straw.
For ten thousand years, every institution and market and power structure on Earth was calibrated to the exact metabolic rate of human speech.
Musk saw the wall before anyone else did.
While the entire industry debates whether AI will take your job, he identified the real extinction event. Not competition. Disconnection.
This is what Neuralink is actually for. Not phone control. Not cursors. Not even paralysis. Those are entry points.
The real project is a bandwidth bridge between carbon and silicon. The only one that could keep humans in the loop before the gap becomes permanent.
Because every network in history has done the same thing. Found its slowest node. And routed around it.
We are about to become the slowest node on the most powerful network ever built.
And one person decided to do something about it.
The danger was never that AI would disagree with us.
When reality’s operating system runs at a trillion bits per second and you are physically capped at three hundred, you don’t get conquered.
You get bypassed.
Language was the technology that separated us from animals. Bandwidth is the technology that will separate us from relevance.
Neuralink isn’t a product. It’s the last bridge off the island before the tide comes in.
English
@TrueAIHound Yes. Catching up fast? Like how radiologists are no longer needed (obviously!) and how nobody will own a car by 2000 because driverless cars are ubiquitous?
English
"we’re already at the point where AI is superhuman in many areas and catching up fast in the rest"
Dude, a free calculator on a smartphone is superhuman. So what?
Can your AI walk into a random kitchen and fix breakfast? Can it clean up the kitchen afterward? If not, it's dumb as a rock. 😀
Just stop belittling the humans that designed and built calculators and your dumb and fake AI.
Haider.@haider1
we’re already at the point where AI is superhuman in many areas and catching up fast in the rest so debating whether to slow or pause AI development from anthropic is unnecessary now when that conversation actually mattered, people were busy mocking LLMs that was a serious failure of judgment
English
@GaryMarcus Never ask a mathematician to explain biology, I guess.
English
Science used to use evidence to make outlandish claims. Here a mathematician is ignoring most of the animal kingdom and leaving only humans and statistical artificial neural networks as ‘intelligent’. Science would also consider the options and full scope of the claim.
To refute the claim of consciousness for LLMs, start with its ‘anatomy’. You can see how the program generates tokens using statistics that are dumped in by a human-training step designed by people. Where is sensory input? Where is the storage of context that overrides the statistics? Why are other apes and animals excluded since they are far more aligned with humans than computer hardware and software!
English
Can we please get this guy into the AI retirement home ASAP.
People that want to believe are going to eat this crap up, but it's a huge embarrassment for the industry, and makes him look like a n00b with psychosis.
Alex Kantrowitz@Kantrowitz
AI Pioneer Geoff Hinton tells me he believes AI is conscious.... and humans better get used to the idea that they're not the only intelligent life on earth. "They've very like us," he says. "They're beings like us." AI chatbots, he says, must understand your questions in order to answer them. There's an awareness there that equates to sentience. "We're going to have to accept that intelligence is not just biological."
English
I blame the tech CEOs for their language claiming that ‘the AI’ did this or that and will soon take our jobs and be worth trillions. Better to ground these programs properly with ‘the statistics made this mistake and that one, and in time its failures will be fixed with good technology’
English
@HedgieMarkets Well, maybe those robots will keep learning and become better soon. I mean, that’s how magic works and the robots could get bored.
English
🦔Picnic, a Seattle pizza robotics startup that raised $53 million and partnered with Domino's, just shut down. The company sold its IP to an unnamed buyer and left at least one restaurant owner stuck with $250,000 of useless robots. Zume Pizza did the same thing in 2023 after burning nearly $500 million trying to keep cheese from sliding off pies inside their delivery trucks. Both companies promised one worker could output 100 pizzas an hour with their hardware.
My Take
Picnic and Zume burned through $550 million combined trying to automate something humans do for $15 an hour. The Seattle restaurant owner stuck with the leftover hardware compared his kitchen to an aquarium of useless machines. That image is funny until you remember the same script is running across Starbucks pulling its AI inventory tool, Waymo pausing eight cities, Microsoft killing Claude Code internally, and Uber blowing through its 2026 AI budget in four months.
The mechanism is always the same in every story I've been covering. The demo works in a controlled environment with clean inputs. The deployment fails because real kitchens, real intersections, and real warehouses produce messy inputs the demo never tested. The vendor gets paid through the failure cycle. The buyer eats the cost and quietly retires the product. If a pizza chef can lose $250,000 on a topping robot, the people writing $80 billion capex checks for general purpose AI agents should expect to learn the same lessons at much larger scale.
Hedgie🤗
English
"our sensors do most of the heavy lifting for the neocortex" - this would mean that bat's sonar does the heavy lifting for them, as vision does for us.
And also human language that can come from almost any sense (vision, touch, hearing, ...). That needs each sense to have the same capabilities, rather than other brain regions like the human neocortex.
My model, Patom theory, uses the strengths of senses, but puts the recognition into the general purpose regions of the cortex as the final arbiters to deal with commonality between senses (so we recognize our spouse by sound if blinded).
Yes, I appreciate this is not aligned with your model.
I like "The world computes itself" other than this removes all computation, since the world does what physics and biology has it do without additional effort! What is cheaper computation than that?
I see brain damage giving a lot of support for the "book keeping system" aligning all our senses with what we experience, and using stored patterns of (simplified) experience to make decisions.
In NLU (Natural Language Understanding) our system aligns the linguistic syntax to semantics and back in context. Both do heavy lifting that I think shows the need for more than just the sensory recognition in our brains.
Anyway, I look forward to your next posts!
English
Yes I believe that our sensors do most of the heavy lifting for the neocortex. The latter is more like a bookkeeping system.
I also believe that most of our intelligence is outside the brain. The world computes itself perfectly while our sensors capture the results in real time, fully computed.
English
Deep learning is irrelevant to solving intelligence. Here's why.
Unlike the brain, a deep neural net cannot perceive an object or pattern unless it has previously been trained to recognize it. In DL, perception is recognition.
The inability of DL systems to perceive objects in the world without recognizing them is the reason that they need so much data. But it's never enough in the real world. Edge cases will invariably pop up and cause a catastrophic failure. This is a fatal flaw if AGI is the goal. It explains why Elon Musk's Tesla has still not solved full self-driving even after a decade of trying.
By contrast, the human brain learns to perceive the world during early childhood. It can see any object or pattern instantly, even if it has not seen it before. I call this ability, "extreme perceptual generalization" or EPG. It's a beautiful thing. There can be no intelligence without it.
Whoever solves EPG essentially solves AGI. The rest is just a walk in the park in comparison. EPG would solve Tesla's pesky edge case problem.
Every time I read a post on X or an article about OpenAI or some other AI lab collecting real-world data for their robotics projects, I can't help laughing. Data collectors are an amusing and pathetic bunch. 😀
I love the brain. I really do. 😍
English
‘This AI race’ isn’t a good description. Whoever wins the race for statistical AI isn’t a solution for human-like capabilities. Real AI is the next generation that aligns with the human brain.
Today’s LLMs are expensive in power and cost. The brain is orders of magnitude cheaper to run and accurate at the same time. Lossy AI isn’t the goal for humanity.
English
Vinod Khosla: "We are in a techno-economic war with China, and we shouldn't call it anything other than a war. Whoever wins this AI race will win the economic race and will win the race for socio-economic power and influence globally."
~ Vinod Khosla, Co-founder of Sun Microsystems & Billionaire Silicon Valley venture capitalist.
---
From 'Fortune Magazine' YT Channel
English
The Turing machine is often cited as the reason computers can do anything, but my brain research shows the fundamental deviation from the computational model to what our brains do.
We live in the computer paradigm age that is slowing progress in AI. Too many false assumptions to deal with causes lazy designs.
Now LLMs demand exponential increases in power to do tasks that only require a laptop. It is bad science and worse engineering!
English
I'm sorry to hear that. I have never found any use for Turing's much hyped "contributions" to my work.
I see no use for Turing in either AI, neuroscience or computer science. Advances in computing, hardware or software, would have continued normally had Turing never existed. He's just a cult figure, imo, and I'm not a member of that cult.
Just a few days ago, speaking to a public audience, Demis Hassabis felt it necessary to mention that the brain was an approximate Turing machine. My thoughts were: Huh? What's the point of that? True or not, who cares? I know I don't.
English
Warning: Not for Turing worshippers.
Turing was the first prompt engineer and the father of fake AI: If it sounds intelligent, it's intelligent. 😀
The Turing machine is a useless academic curiosity mostly used for showing off. Computers are not Turing machines. Sci-fi fruitcakes and the fake AI mafia love to mention Turing for some reason. 😮
Then atheist fruitcake Richard Dawkins showed up and pushed the Turing test further into weirdo-land: If Claudia sounds conscious, she's probably conscious. 😀
Pedro Domingos@pmddomingos
Turing wasn't Turing-complete.
English
@TrueAIHound If you’re not designing code that replicates human capabilities I’d imagine not much use for Turing’s work. My point was that people thinking the imitation game is just conversational chat haven’t understood what he designed.
Again, I’m not a fan of those behaviourist tests.
English
Well, in 2023 AGI was really close too. Better - save predictions until something like it already can be done. Otherwise it can look like hype to drive stock price instead of being a scientific claim.
Geoffrey Hinton predicted the demise of radiologists in 2016 and that only caused an under supply of them. And driverless cars due by 2020 didn’t happen either.
English
Demis Hassabis says AGI is now only a few years away, and 'singularity' is the right word for the era it could create
A technology this transformative makes predictions beyond it almost impossible
"looking back from 5 to 10 years later, 2026–27 will be seen as when it was starting"
English
@realBigBrainAI Or at least stop those calling statistical programs AI and only call AI things that are ‘intelligent’ or that at some level duplicate cognition.
English
Roger Penrose, Nobel Prize-winning physicist and mathematician, explains why we should stop calling it AI and start calling it "artificial cleverness":
He believes the entire field is mislabelled, and the label itself is doing damage.
His objection is simple but cuts deep:
"The name is wrong. It's not artificial intelligence. It's not intelligence. Intelligence would involve consciousness. Well, if it's a machine, it's not conscious."
For Penrose, people have confused raw computing power with genuine understanding.
"People have lost the plot. They've lost it in the power of computing. The thing is that computers have got so powerful that they've lost the thread of what they're doing. But I think consciousness is something different. It's not computational."
He believes the term itself has hypnotized people into a category error:
"People are so hypnotized. The trouble is that AI is a bad term. It means artificial intelligence. Now intelligence in my view is conscious. That's what intelligence is about."
So he proposes a rename. Artificial Cleverness. AC instead of AI.
To illustrate the distinction, Penrose draws on his experience teaching mathematics:
"You have mathematics students. Some of them understand what they're doing. Some are just clever. They can repeat what they've learned. They know how to do it very cleverly. They can calculate very well, but they don't necessarily understand what they're doing."
That gap, between calculating well and actually understanding, is the gap Penrose sees between today's machines and genuine intelligence.
Cleverness can be manufactured. Consciousness, in his view, cannot.
So the question worth sitting with: when we call a system "intelligent," are we describing what it does, or quietly assuming something about what it is?
English