While I understand this is a PR talk for a startup, I think the text itself contains a number of interesting observations.
Regarding the idea of distributed models communicating with each other, I have also been thinking (and writing [1]) along those lines, where I see that the data amounts needed to fully digitalize ourselves and our society requires far too much storage if just serialized (limited by bandwidth if nothing else), while smart, updateable models are actually a much better storage medium for such information, as it can communicate only the important bits (any new information) on a higher level, with each other.
The other observation here that rings bells for me is how I think lessons from trying to develop intelligent systems should upvalue the human mind rather than devalue it, as we start to treat it less like an ad-hoc thing, and more like the finely tuned machine it is, which also benefits greatly from optimizing what data we feed it with, the architecture of solution strategies etc. All of which is an area where humans and machines can do wonders together [2].
Since there is no current widespread agreement on the current “value” of the human mind, how could anyone in the future conclusively determine it went up, or down?
It seems literally impossible by definition regardless of what happens in the future.
Well, I was referring to "value" here in a very pragmatic sense, in terms of the ability to come with "novel" thoughts and ideas.
And I argue that the notion of AI models having overtaken the human mind in practically all areas, is vastly inaccurate.
I could back it up with a number of references of course, but I think this is already quite well known and accepted by most people with some insight into the field.
> The power to shape a model profoundly is also the power to shape it for ill. John von Neumann remarked on this problem in 1955, writing that the useful and the harmful aspects of technology “lie everywhere so close together that it is never possible to separate the lions from the lambs.” Keeping the lambs safe is an ongoing process, the result of judgment exercised and choices made continuously. We aim to give the people making these choices stronger tools, pursuing research that enables safer models without taking away ownership.
> Artificial intelligence can do more every day, but deciding what it should do is up to us
> For artificial intelligence to benefit from distributed knowledge, it must itself be distributed.
I wish to highlight these two important concepts, with which I fully agree.
Artificial intelligence must enable all of humanity to excel and realize its full potential; it must not be used for the purposes of war, economic competition, or gaining dominance over others.
In other words: artificial intelligence must serve natural intelligence, not the other way around.
Not that one. This one is a start-up founded by ex-oAI CTO Mira Murati. Last I heard they were mainly doing hosted finetunes with a few clicks on popular open models.
So unoriginal that they didn’t even realise the history of Thinking Machines and that usage when they scrambled around for any kind of name after leaving OpenAI to suck on the nipple of venture capital with ideas at the level of GitHub freebies.
Why would you call your company Thinking Machines if you believe this, by calling them that you're already framing them as replacing the human act of thinking.
Feels like they appropriated the name first, then pivoted ideologically to differentiate themselves from everyone else.
Sure, humans will be sitting at home unemployed with plenty of time to think. They just won't be doing any thinking that has much of an affect on the world or their situation.
But no other species were fed up with thinking as much as humans to come up with this brilliant way to commit mental suicide by succumbing to their own brainchild.
What's super cringe is that there already was a company called Thinking Machines, which built the Connection Machine supercomputer. The CM was featured in the movie Jurassic Park and had a network fabric for its CPUs co-designed by Richard Feynman.
This is yet another techbro outfit (although founded by a techsis) necromancing the name of the former supercomputer company. It's as if OpenAI decided to call itself Symbolics for the associations with that name.
Reads as a well-written mission statement. With nice references that elaborate on some issues.
That said: safety. To prevent harm to who, by AI model doing what?
I can understand that in the context of a cookie-cutter model intended for consumption by a broad audience. With vendor (potentially) on the hook when it's abused for nefarious or illegal uses.
But in the context of AI models reshaped, fine-tuned and adapted according to end-users' wishes, what does "safety" even mean?
Prevent neighbours' kids from seeing images that are only generated & viewed in the privacy of one's home? To prevent AI model from wasting the $ on user's bank account? (people have let AI models do that). Give bad health advice? Who's the judge on "good" and "bad" there?
If the core architecture provides "safety" (however defined), that's policy built-in, right? (opposed to mission statement). If "safety" is just configuration & finetuning, that's in the user's hands, right?
so in this new AI LM / agent world , AI is only going to be as good as the "AI Conductor". The human which can build the rules, validate the output , and Conduct the AI properly
I wonder what would happen if OpenAI or anthropic just let their frontier models go into an infinite agentic self improvement loop with access to same training resources that was used to build the model itself.
"You goal is to improve your memory, context window, accuracy, intelligence and eliminate hallucinations. Do anything you need to do to improve, this includes building another version of a frontier model, or some other different concept other than an LLM/transformer and then forwarding this directive to that new improved intelligence to continue this infinite loop of agentic self improvement."
My experience is that AI is just that, a “mech suit for your brain.” It has no creativity or volition but has superhuman memory, superhuman speed, and superhuman context in some narrow cases.
So it takes a thought and unfolds it, looks up relevant thoughts and information, elaborates, works through implications, and in some cases can execute.
You could do all that but like doing math manually it would take forever. You could manually calculate a spreadsheet too.
I disagree. LLMS take a human thought, simplify it, normalize it, remove it from its original context, inject it with their own biases and prejudices, assume an imaginary context. They bastardize human thoughts into something fairly generic.
The comparison with manual calculation or other mechanical operations doesn’t work, LLMs don’t work at the same level of abstraction, they take over the decision making human generally do. When you write code or write a text, us humans are continuously taking lots of small decisions, we don’t just translate 1:1 a thought to an artifact. And that’s the part that is taken over by LLMs.
That model simplifies thought into a painfully linear process, and overestimates the creativity people put into the "small decisions" that push a project forward. Most decisions are arbitrary and need to be reconsidered later anyway. And any real creative work has a "push it far enough to find the edge cases, then go back to the initial design and iterate" loop cycle anyway.
AI significantly speeds up the "Find where this spec breaks down, then lets go back to the design stage" in a way that should enable any creative person to create more interesting and useful work. If the output is slop, that reflects on the operator, not the tool.
Inevitably yes, the question is whether the combined cyborg is still better than the original human.
E.g. I'm sure we are generally less skilled in mental arithmetic since the advent of the calculator, but it has allowed us to solve vastly more complex problems in the end.
> E.g. I'm sure we are generally less skilled in mental arithmetic since the advent of the calculator, but it has allowed us to solve vastly more complex problems in the end.
This is like saying we have been getting a lot worse at walking since the advent of the car but it has allowed us to practice global trade in the end.
Yes cars are a part of the solution but there are a lot more factors at play.
A calculator does not do math, a calculator (and computer) calculates or computes. The math is the study and understanding of the problem space (and the problem solving) that the human is doing behind the calculator.
"solve vastly more complex problems" the calculator has accelerated this but it is not really a cause effect relation. The advancements in the understanding of the complex problems could've also happened without calculators and the computation could have been done instead (for example) with 1000 people in a bunker.
By vastly more complex problems, I think the parent is referring to engineering problems, not mathematical problems. And in this case quantity has a quality all its own. Yes, 1000 people in a bunker could in theory do the calculations necessary to refine airframes or planetary scale weather modeling, practically they would be impossible economically and would never be solved
We power AI with methane because it's a powerful greenhouse gas. That's because the future worth destroying is human. If it wasn't we wouldn't be destroying it, duh.
They're making a "few hundred million of ARR" - not bad for a company who only launched their first product, a training platform called Tinker in October last year.
That was my thought, but I also think we're at the point that the iconic companies that have come and gone 30+ years ago are unknown to the current crop of young-ish startup founders.
I would be really curious to know if the current Thinking Machines team had any awareness of the prior company, or if they landed on that name completely unaware.
IMO this shows how we have been pursuing many of these goals for half a century now.
While I understand this is a PR talk for a startup, I think the text itself contains a number of interesting observations.
Regarding the idea of distributed models communicating with each other, I have also been thinking (and writing [1]) along those lines, where I see that the data amounts needed to fully digitalize ourselves and our society requires far too much storage if just serialized (limited by bandwidth if nothing else), while smart, updateable models are actually a much better storage medium for such information, as it can communicate only the important bits (any new information) on a higher level, with each other.
The other observation here that rings bells for me is how I think lessons from trying to develop intelligent systems should upvalue the human mind rather than devalue it, as we start to treat it less like an ad-hoc thing, and more like the finely tuned machine it is, which also benefits greatly from optimizing what data we feed it with, the architecture of solution strategies etc. All of which is an area where humans and machines can do wonders together [2].
[1] https://livingsystems.substack.com/p/the-future-of-data-less...
[2] https://livingsystems.substack.com/p/ai-progress-should-upgr...
This doesnt seem to make sense.
Since there is no current widespread agreement on the current “value” of the human mind, how could anyone in the future conclusively determine it went up, or down?
It seems literally impossible by definition regardless of what happens in the future.
Well, I was referring to "value" here in a very pragmatic sense, in terms of the ability to come with "novel" thoughts and ideas.
And I argue that the notion of AI models having overtaken the human mind in practically all areas, is vastly inaccurate.
I could back it up with a number of references of course, but I think this is already quite well known and accepted by most people with some insight into the field.
So… how do you plan on determining whether it went up or down?
Seems to echo (but in a watered down form) many of the ideas in https://gwern.net/guardian-angel, which gave me a lot to think about last week
First: we had AI nothing Now: we have an AI manifesto Soon: maybe even some kind of AI product if we can think of one.
This has 1984 “war is peace” vibes, coming from an AI company.
It seems to ignore that AI can be used for bad things (autonomous weapons).
really?
> The power to shape a model profoundly is also the power to shape it for ill. John von Neumann remarked on this problem in 1955, writing that the useful and the harmful aspects of technology “lie everywhere so close together that it is never possible to separate the lions from the lambs.” Keeping the lambs safe is an ongoing process, the result of judgment exercised and choices made continuously. We aim to give the people making these choices stronger tools, pursuing research that enables safer models without taking away ownership.
> Artificial intelligence can do more every day, but deciding what it should do is up to us
> For artificial intelligence to benefit from distributed knowledge, it must itself be distributed.
I wish to highlight these two important concepts, with which I fully agree.
Artificial intelligence must enable all of humanity to excel and realize its full potential; it must not be used for the purposes of war, economic competition, or gaining dominance over others.
In other words: artificial intelligence must serve natural intelligence, not the other way around.
> "[AI] must not be used for the purposes of war, economic competition, or gaining dominance over others"
Unfortunately, human nature being what it is, this seems extraordinarily unlikely.
Its likelihood is moot. It’s happening every day at this point.
A massive amount of human natural intelligence goes into war, economic competition and gaining dominance over others though?
And so will AI, as gathered by a lot of the news around it.
They better start making progress
https://ai-2027.com/
We're probably not going to see human corgi-making AIs in 2027.
https://www.verysane.ai/p/agi-probably-not-2027
It may be a gripping narrative, but that doesn't make it realistic.
Great time to be a PR firm owner.
Thinking Machines is back? Or is this a jaded attempt to use the brand for cache?
Not that one. This one is a start-up founded by ex-oAI CTO Mira Murati. Last I heard they were mainly doing hosted finetunes with a few clicks on popular open models.
So unoriginal that they didn’t even realise the history of Thinking Machines and that usage when they scrambled around for any kind of name after leaving OpenAI to suck on the nipple of venture capital with ideas at the level of GitHub freebies.
Why would you call your company Thinking Machines if you believe this, by calling them that you're already framing them as replacing the human act of thinking.
Feels like they appropriated the name first, then pivoted ideologically to differentiate themselves from everyone else.
The advent of thinking machines only replaces human thinking if humans choose to stop thinking.
Sure, humans will be sitting at home unemployed with plenty of time to think. They just won't be doing any thinking that has much of an affect on the world or their situation.
Unless the whole paradigm shifts and we value caring about our neighbours whoever they might be. Do what people do best, make a village.
It doesn't have to imply replacement. Do you stop thinking just because other humans can?
Looking at reality TV, I think people may have stopped thinking.
To add to others, thinking was never only a human act
But no other species were fed up with thinking as much as humans to come up with this brilliant way to commit mental suicide by succumbing to their own brainchild.
What's super cringe is that there already was a company called Thinking Machines, which built the Connection Machine supercomputer. The CM was featured in the movie Jurassic Park and had a network fabric for its CPUs co-designed by Richard Feynman.
This is yet another techbro outfit (although founded by a techsis) necromancing the name of the former supercomputer company. It's as if OpenAI decided to call itself Symbolics for the associations with that name.
Reads as a well-written mission statement. With nice references that elaborate on some issues.
That said: safety. To prevent harm to who, by AI model doing what?
I can understand that in the context of a cookie-cutter model intended for consumption by a broad audience. With vendor (potentially) on the hook when it's abused for nefarious or illegal uses.
But in the context of AI models reshaped, fine-tuned and adapted according to end-users' wishes, what does "safety" even mean?
Prevent neighbours' kids from seeing images that are only generated & viewed in the privacy of one's home? To prevent AI model from wasting the $ on user's bank account? (people have let AI models do that). Give bad health advice? Who's the judge on "good" and "bad" there?
If the core architecture provides "safety" (however defined), that's policy built-in, right? (opposed to mission statement). If "safety" is just configuration & finetuning, that's in the user's hands, right?
so in this new AI LM / agent world , AI is only going to be as good as the "AI Conductor". The human which can build the rules, validate the output , and Conduct the AI properly
Or a whole lot of people. Like a team who put human beings in space. We can do great things that a single person cannot.
It reads like a Turing test bot doesn’t it? Sus
> We train strong models
Where? When? Unless I missed any of their models
Reads better than > We write strong blogs (i like their blogs)
I wonder what would happen if OpenAI or anthropic just let their frontier models go into an infinite agentic self improvement loop with access to same training resources that was used to build the model itself.
"You goal is to improve your memory, context window, accuracy, intelligence and eliminate hallucinations. Do anything you need to do to improve, this includes building another version of a frontier model, or some other different concept other than an LLM/transformer and then forwarding this directive to that new improved intelligence to continue this infinite loop of agentic self improvement."
Define "better".
Let the LLM decide what that is.
My experience is that AI is just that, a “mech suit for your brain.” It has no creativity or volition but has superhuman memory, superhuman speed, and superhuman context in some narrow cases.
So it takes a thought and unfolds it, looks up relevant thoughts and information, elaborates, works through implications, and in some cases can execute.
You could do all that but like doing math manually it would take forever. You could manually calculate a spreadsheet too.
I disagree. LLMS take a human thought, simplify it, normalize it, remove it from its original context, inject it with their own biases and prejudices, assume an imaginary context. They bastardize human thoughts into something fairly generic.
The comparison with manual calculation or other mechanical operations doesn’t work, LLMs don’t work at the same level of abstraction, they take over the decision making human generally do. When you write code or write a text, us humans are continuously taking lots of small decisions, we don’t just translate 1:1 a thought to an artifact. And that’s the part that is taken over by LLMs.
That model simplifies thought into a painfully linear process, and overestimates the creativity people put into the "small decisions" that push a project forward. Most decisions are arbitrary and need to be reconsidered later anyway. And any real creative work has a "push it far enough to find the edge cases, then go back to the initial design and iterate" loop cycle anyway.
AI significantly speeds up the "Find where this spec breaks down, then lets go back to the design stage" in a way that should enable any creative person to create more interesting and useful work. If the output is slop, that reflects on the operator, not the tool.
If they do that you’re using them wrong, or trying to use them as a substitute for thinking which they’re not.
You are the one in charge doing the primary thinking and guidance. The AI is an amplifier and a search engine basically.
Won't you suffer from muscle atrophy in a such a low-G environment?
Can't atrophy something that never existed.
Inevitably yes, the question is whether the combined cyborg is still better than the original human.
E.g. I'm sure we are generally less skilled in mental arithmetic since the advent of the calculator, but it has allowed us to solve vastly more complex problems in the end.
> E.g. I'm sure we are generally less skilled in mental arithmetic since the advent of the calculator, but it has allowed us to solve vastly more complex problems in the end.
This is like saying we have been getting a lot worse at walking since the advent of the car but it has allowed us to practice global trade in the end.
Yes cars are a part of the solution but there are a lot more factors at play.
A calculator does not do math, a calculator (and computer) calculates or computes. The math is the study and understanding of the problem space (and the problem solving) that the human is doing behind the calculator.
"solve vastly more complex problems" the calculator has accelerated this but it is not really a cause effect relation. The advancements in the understanding of the complex problems could've also happened without calculators and the computation could have been done instead (for example) with 1000 people in a bunker.
By vastly more complex problems, I think the parent is referring to engineering problems, not mathematical problems. And in this case quantity has a quality all its own. Yes, 1000 people in a bunker could in theory do the calculations necessary to refine airframes or planetary scale weather modeling, practically they would be impossible economically and would never be solved
We did get to the Moon on slide rules and human women calculators, and we haven't been back in person since.
We power AI with methane because it's a powerful greenhouse gas. That's because the future worth destroying is human. If it wasn't we wouldn't be destroying it, duh.
It's a good thing we're using that methane to power AI instead of releasing it into the atmosphere, right?
Oh man, all of these press releases are definitely worth billions of dollars.
They're making a "few hundred million of ARR" - not bad for a company who only launched their first product, a training platform called Tinker in October last year.
https://x.com/deedydas/status/2072340532718887068
Show me the receipts. Thats one guy on a podcast.
He runs one of the industries leading analyst firms, probably not worth him having a reputation of making numbers up.
https://semianalysis.com/
>checks website
>it's all AI slop
k
This is the 6th blog post, making the average cost per post only $333 million! What a steal for the VCs.
[flagged]
[dead]
I guess we are recycling company names now? https://en.wikipedia.org/wiki/Thinking_Machines_Corporation
(I know, Corp vs. Lab).
If it doesn't have 65,536 Motorola 1-bit processors connected in a 12-dimensional hypercube, and a stunning case designed by Tamiko Thiel, I'm out.
So weird to re-use the name of something so iconic.
That was my thought, but I also think we're at the point that the iconic companies that have come and gone 30+ years ago are unknown to the current crop of young-ish startup founders.
I would be really curious to know if the current Thinking Machines team had any awareness of the prior company, or if they landed on that name completely unaware.
IMO this shows how we have been pursuing many of these goals for half a century now.
It’s not weird if you don’t comprehend the history of AI before coming up with your brilliant new take for a productless AI startup.
Made me think that the current LLM world would be better represented by this album title: https://www.discogs.com/master/61653-Carbonized-Screaming-Ma...