I can tell you for a fact that they can. However, even managing boilerplate and repetitive code is a huge benefit. Furthermore, these tools are great at combing through code bases and helping you find where you need to make changes in code. If you haven’t actually used these tools in a real project yourself then you don’t really know what they’re capable of.

Indeed, self hosing has to be the way forward.

It depends on the task and the specific LLM. My experience is that they can do a lot of things effectively nowadays, and they’re improving rapidly.

seems fine for me, here’s the content:

Mainland China is on track to surpass Taiwan in semiconductor foundry capacity by 2030, according to a report from Yole Group, underscoring Beijing’s progress in its push for chip self-sufficiency amid ongoing US tech restrictions. The mainland’s share of global foundry capacity is projected to reach 30 per cent by the end of the decade, up from 21 per cent in 2024, the French market research firm said. Taiwan is currently the market leader with a 23 per cent share last year, while mainland China is already ahead of South Korea at 19 per cent, Japan at 13 per cent and the US at 10 per cent. “Mainland China is rapidly becoming a central player,” Yole Group said, attributing the shift to Beijing’s intensified efforts to build a self-sufficient domestic semiconductor ecosystem since Washington launched a tech war that aimed to curb China’s progress in critical areas such as chips and artificial intelligence (AI). Beijing has doubled down on its “whole nation” approach to its self-sufficiency drive. The state-backed China Integrated Circuit Industry Investment Fund, known as the “Big Fund”, has successfully fostered the development of key companies such as Semiconductor Manufacturing International Corporation (SMIC) and Hua Hong Semiconductor, two of the country’s leading wafer foundries. Domestic fabs are set to play a bigger role over the next few years, according to the report, which said local chipmakers accounted for 15 per cent of foundry capacity in 2024. That share will be “significantly more” by 2030, the report said. Chinese chipmakers have been investing heavily in expanding their facilities to meet surging demand from sectors such as automotive and generative AI. China was expected to start three new fab construction projects this year, one-sixth of the world’s total, according to a report published in January by US-based industry association SEMI. China’s self-sufficiency strategy, along with expected demand from automotive and internet-of-things applications, would help boost capacity by 6 per cent for chips made with process nodes between 8 and 45 nanometres, SEMI added. Despite the projected gains, the mainland still trails Taiwan and South Korea in advanced process nodes, which are crucial for producing high-performance chips with greater transistor density. SMIC, China’s top foundry, had difficulty advancing its process nodes from 7-nm to 5-nm, Canadian research firm TechInsights said in a report last month. Two years after its 7-nm chip first appeared in a Huawei Technologies smartphone, “SMIC’s 5nm process node remains elusive,” TechInsights said. The report came after it looked into the chip used in Huawei’s new laptop with a foldable display, which also used 7-nm chips from SMIC. Meanwhile, global leaders Taiwan Semiconductor Manufacturing Company (TSMC) and Samsung Electronics are locked in a race to achieve mass production at the 2-nm node level. TSMC was expected to reach that level this year, while Samsung has reportedly planned to reach the same stage in early 2026.

I think that’s exactly what’s gonna happen in the long run. Right now we’re in the hype phase of a new technology, but one the hype dies down we’ll start identifying use cases where the tech actually works well. At the same time the tech itself is going to mature, and people will figure out how to work with it effectively.

Yeah, like how do they not see this project as being strategic for them.

related research

• https://www.sciencedirect.com/science/article/pii/S2666498425000031
• https://www.sciencedirect.com/science/article/pii/S2590332221002888

Have some humility and willingness to learn.

I have plenty of willingness to learn from people who have a clue on the subject.

I didn’t say it was the primary function.

You literally tried to argue that evolution doesn’t create complexity if there’s a more efficient path.th.

Then what about Darwin who literally said, “Natural selection is continually trying to economize every part of the organization.” Now please go and read some introductory texts on biology before trying to explain to me why Darwin is wrong. There’s so much going on when it comes to the thermodynamics of living systems and you’re clearly not ready to have a conversation about it.

Again, you’re showing a superficial understanding of the subject here. Natural selection selects for overall fitness, and efficiency is only a small part of equation. For example, plants don’t use the most efficient wavelength for producing energy, they use the one that’s most reliably available. Similarly, living organisms have all kinds of redundancies that allow them to continue to function when they’re damaged. Evolution optimizes for survival over efficiency.

You’re baseless assuming that hydrocephalus causes the brain to lose a substantial amount of its complexity.

Maybe read the actual paper linked there?

But hey neuroscience hasn’t really advanced at all since 1980 right? The brain is totally redundant right? There’s no possible way a critical and discerning person such as yourself could have been taken in by junk science, right?!!

What I linked you is a case study of an actual living person who was missing large parts of their brain and had a relatively normal life. But hey why focus on the actual facts when you can just write more word salad right?

I took issue with specific statements you made that stand apart from the rest of your comment.

You took issue with made up straw man arguments that you yourself made and have fuck all with what I actually said. Then you proceeded to demonstrate that you don’t actually understand the subject you’re debating. You might as well start believing in the astrology, crystals, and energy healing. At least those interests will make you seem fun and quirky instead of just a sad debate bro.

Im simply stating that you’re way off base when you claim that they appear to operate using the same principles or that all evidence suggests the human mind is nothing more than a probability machine.

I literally said these things, and you never gave any actual counter argument to either of them.

You’re betraying your own ignorance about neuroscience. The complexity of the brain is absolutely linked with its ability to reason and we have plenty of evidence to show that. The evolutionary process does not just create needless complexity if there is a more efficient path.

You’re betraying your ignorance of how biology works and illustrating that you have absolutely no business debating this subject. Efficiency is not the primary fitness function for evolution, it’s survivability. And that means having a lot of redundancy baked into the system. Here’s a concrete example for you of just how much of the brain isn’t actually essential for normal day to day function. https://www.rifters.com/crawl/?p=6116

This is such a silly statement especially when you’ve been claiming that both the brain and AI appear to work using the same principles.

There’s nothing silly in stating that the underlying principles are similar, but we don’t understand a lot of the mechanics of the brain. If you truly can’t understand such basic things there’s little point trying to have a meaningful discussion.

I don’t really care about your arguments concerning embodiment because they’re so beside the point when you just blowing right by the most basic principles of neuroscience.

That’s literally the whole context for this thread, it just doesn’t fit with the straw man you want to argue about.

A ruthless criticism of that exists includes the very researchers whose work you’re taking at face value.

Whose work am I taking at face value specifically? You’re just spewing nonsense here without engaging with anything I’m saying.

I suspect that something like LLMs is part of our toolkit, but I agree that this can’t be the whole picture. Ideas like neurosymbolic AI might be on the right track here. The idea here is to leverage LLMs at parsing and classifying noisy input data, which they’re good at, then use a symbolic logic engine to operate on the classified data. Something along these lines is much more likely to produce genuine intelligence. We’re still in very early stages of both understanding how the brain works and figuring out how to implement artificial reasoning.

LLMs and the human mind operate on categorically different principles.

A bold statement given that we don’t actually understand how the brain operates exactly and what algorithms that would translate into.

Where the straw man?

The straw man is you continuing to argue against equating LLMs with the functioning of the brain, something I never said here.

All the verbiage used to describe neural network models has little to do with how the brain actually works.

You appear to be conflating the implementation details of how the brain works with the what it’s doing in a semantic sense. There is zero evidence that all the complexity of the brain is inherent to the way our reasoning functions. Again, we don’t have a full understanding of how the brain accomplishes tasks like reasoning. It may be a lot more complex than what LLMs do, or it may not be. We do not know.

Finally, none of this has anything to do with the point I was actually making which is regarding embodiment. You decided to ignore that to focus on braying about tech companies and LLMs instead.

This completely understates the gulf between what we call AI and how the human brain actually works.

Way to completely misrepresent what I was actually saying. Nowhere was I suggesting that there isn’t a huge difference between the two. What I pointed out is that, while undeniably more complex, our brains appear to work on similar principles.

My only point was that the feedback loop from embodiment creates the basis for volition, and that what we call intelligence is our ability to create internal models of the world that we use for decision making. So, this is likely a prerequisite for any artificial system that has any meaningful intelligence.

Maybe try engaging with that instead of writing a wall of text arguing with a straw man.

All the evidence suggests that our own minds are also nothing more than probability engines. The reason we consider humans to be intelligent is because our brains learn to model the events in the physical world that are fed into our brains by the nervous system. The whole purpose of a brain is to try and keep the body in a state of homeostasis. That’s the basis for our volition. The brain gets data about about the state of the organism, and interprets it as hunger, pain, fear, and so on. Then it uses its internal world model to figure out actions that will put the body into a more desirable state. From this perspective, embodiment would indeed be a necessary component of human style intelligence.

While LLMs on their own are unlikely to provide a sufficient basis for a reasoning system, its not strictly impossible that a model trained on sensory data from a robot body it inhabits wouldn’t be able to build a representation of the world and its body that could be used as the basis for decision making and volition.

I thought the game was alright overall, but it certainly did not feel like a Dragon Age game. The overall story was decent, however a lot of the dialogue was hamfisted. The real problem was that the gameplay felt like Jedi Survivor without the refined combat mechanics. As a result, combat quickly became tedious and repetitive. I also found that the NPCs were more or less fungible, and it really didn’t matter who was in your party. This is a stark contrast with previous Dragon Age games where the whole fun was in scripting the behavior of different characters and coming up with clever ways of combine their abilities. Simply having kept the original mechanics, warts and all, would’ve resulted in a far better game in my opinion.

US threats mean nothing because the US has no bargaining power, and everybody knows this now.

Human-like object concept representations emerge from datasets made by humans because humans made them.

And humans made them that way because human minds evolved to represent data in this way. As I keep pointing out, we’re feeding data into neural networks that’s organized in a way that’s natural for our brains to operate on. It’s an artificial system that mimics the way we naturally represent data in our own minds.

The artificial aspect of the system lies in the implementation details. The ways we’ve come up to encode data. These are not essential. It’s like a difference between an algorithm, and its concrete implementation in a programming language. The fact that the data is encoded using human designed formats is incidental to the structure of the data which is derived from the way our brains encode information.

Human-like object concept representations emerge from the way our brains are structured. These are the representations that are encoded into data sets by humans.

Also, you’ve talked about a dialectical relationship, but dialectics are about understanding evolution of dynamic systems. The contradictions represent the opposing forces within a system that guide its development over time. When we talk about a distinction between natural and artificial, what’s the system that we’re discussing here what are the opposing forces?

I haven’t defined artificial out of existence at all. My definition of artificial is a system that was consciously engineered by humans. The human mind is a product of natural evolutionary processes. Therefore, the way we perceive and interpret the world is inherently a natural process. I don’t see how it makes sense to say that human representation of the world is not natural.

An example of something that’s artificial would be taking a neural network we designed, and having it build a novel representation of the world that’s unbiased by us from raw inputs. It would be an designed system, as opposed to one that evolved naturally, with its own artificial representation of the world.

Are you saying that humans are not a product of nature?

You continue to ignore my point that human representation are themselves not arbitrary. Our brains have emerged naturally, and that’s what makes the representations humans make natural. You could evolve a representation of the model from scratch by hooking up a neural network to raw sensory inputs, and its topology will eventually become tuned to model those inputs. I don’t see what would be fundamentally more natural about that though.

A more accurate conclusion would be: human-like object concept representations emerge when fed data collected by humans, curated by humans, annotated by humans, and then tested by representation learning methods designed for humans.

Again, I’m not disputing this point, but I don’t see why it’s significant to be honest. As I’ve noted, human representation of the world is not arbitrary. We evolved to create efficient models that allow us to interact with the world in an effective way. We’re now seeing that artificial neural networks are able to create similar types of internal representations that allow them to meaningfully interact with the data organized in a way that’s natural for humans.

I’m not suggesting that human style representation of the world is the one true way to build a world model, or that other efficient representations aren’t possible. However, that in no way detracts from the fact that LLMs can create a useful representation of the world, that’s similar to our own.

Ultimately, the end goal of this technology is to be able to interact with humans, to navigate human environments, and to accomplish tasks that humans want to accomplish.

It’s a good thing in a sense that it means the models are creating stable representations of objects across modalities. It means that there is potential for extending LLM approach to building actual world models in the future.

It’s not merely natural. It’s human.

I’m not disputing this, but I also don’t see why that’s important. It’s a representation of the world encoded in a human format. We’re basically skipping a step of evolving a way to encode this data.

We know that LLMs, when fed human-like inputs, produce human-like outputs. That’s it. That tells us more about LLMs and humans than it tells us about nature itself.

Did you actually read through the paper?

I didn’t say they’re encoding raw data from nature. I said they’re learning to interpret multimodal representations of the encodings of nature that we feed them in human compatible formats. What these networks are learning is to make associations between visual, auditory, tactile, and text representations of objects. When a model recognizes a particular modality such as a sound, it can then infer that it may be associated with a particular visual object, and so on.

Meanwhile, the human perspective itself isn’t arbitrary either. It’s a result of evolutionary selection process that shaped the way our brains are structured. This is similar to how brains of other animals encode reality as well. If you evolved a neural network on raw data from the environment, it would eventually start creating similar types of representations as well because it’s an efficient way to model the world.