If it's not reprogrammable, it's just expensive glass.
If you etch the bits into silicon, you then have to accommodate the bits by physical area, which is the transistor density for whatever modern process they use. This will give you a lower bound for the size of the wafers.
This can give huge wafers for a very set model which is old by the time it is finalized.
Etching generic functions used in ML and common fused kernels would seem much more viable as they could be used as building blocks.
Models don’t get old as fast as they used to. A lot of the improvements seem to go into making the models more efficient, or the infrastructure around the models. If newer models mainly compete on efficiency it means you can run older models for longer on more efficient hardware while staying competitive.
If power costs are significantly lower, they can pay for themselves by the time they are outdated. It also means you can run more instances of a model in one datacenter, and that seems to be a big challenge these days: simply building an enough data centres and getting power to them. (See the ridiculous plans for building data centres in space)
A huge part of the cost with making chips is the masks. The transistor masks are expensive. Metal masks less so.
I figure they will eventually freeze the transistor layer and use metal masks to reconfigure the chips when the new models come out. That should further lower costs.
I don’t really know if this makes sanse. Depends on whether we get new breakthroughs in LLM architecture or not. It’s a gamble essentially. But honestly, so is buying nvidia blackwell chips for inference. I could see them getting uneconomical very quickly if any of the alternative inference optimised hardware pans out
From my own experience, models are at the tipping point for being useful at prototypes in software, and those are very large frontier models not feasible to get down on wafers unless someone does something smart.
I really don't like the hallucination rate for most models but it is improving, so that is still far in the future.
What I could see though, is if the whole unit they made would be power efficient enough to run on a robotics platform for human computer interaction.
It makes sense they would try to make repurposing their tech as much as they could since making changes is frought with a long time frame and risk.
But if we look long term and pretend that they get it to work, they just need to stay afloat until better smaller models can be made with their technology, so it becomes a waiting game for investors and a risk assessment.
> From my own experience, models are at the tipping point for being useful at prototypes in software
You must not have much experience using the new frontier models then. A lot of large tech companies are replacing their SDLC with agentic workflows. The tooling and frameworks are still ramping up, but the models have no problem producing production ready software given proper specifications.
Reading the in depth article also linked in this thread, they say that only 2 layers need to change most of the time. They claim from new model to PCB in 2 months. Let's see, but sounds promising.
If you etch the bits into silicon, you then have to accommodate the bits by physical area, which is the transistor density for whatever modern process they use. This will give you a lower bound for the size of the wafers.
And Unix was mainly made by two people, it's astounding that as I get older, even tech managers don't know "the mythical man month", and how software production generally scales.
Thanks, I learned something, but the original point stands, 5 people is still not a lot and well within the scale where you could manage things within the team yourself without dedicated management and have first hand information flow.
I do agree with this idea in the sense that companies keep trying to add people to projects to do more things or complete projects sooner which ends up wasting a lot of effort. A more cost conscious way is to have smaller teams and let them more time to explore better approaches for longer.
> Sorry but a 99.999% of developers could not have built Unix. Or Winamp.
> Managers are crossing their fingers that devs they hire are no worse than average, and average isn't very good.
The problem is that that's the same skill required to safely use AI tools. You need to essentially audit its output, ensure that you have a sensible and consistent design (either supplied as input or created by the AI itself), and 'refine' the prompts as needed.
AI does not make poor engineers produce better code. It does make poor engineers produce better-looking code, which is incredibly dangerous. But ultimately, considering the amount of code written by average engineers out there, it actually makes perfect sense for AI to be an average engineer — after all, that's the bulk of what it was trained on! Luckily, there's some selection effect there since good work propagates more, but that's a limited bias at best.
Agree completely. Where I'm optimistic about AI is that it can also help identify poorly written code (even it's own code), and it can help rewrite it to be better quality. Average developers can't do this part.
From what I've found it's very easy to ask the AI to look at code and suggest how to make the code maintainable (look for SRP violations, etc, etc). And it will go to work. Which means that we can already build this "quality" into the initial output via agent workflows.
I'm looking forward to the day where an ml/functional inspired language can be used for real time rendering and game engines, how far are we from that?
Realistically, one could argue it's not the right choice overall, but still, it's an application which would push the boundaries of what those languages have been perceived to have the greatness weakness in. An application which is mostly about handing mutable state with high performance.
Rust is a good candidate, but it lacks some crucial aspects when it comes to what I would consider 'nice to haves' from a modern language in this territory.
While rust has traits, borrowing etc, it doesn't have a lot of things with regard to types and optimization. Things like:
- A lack of GADTs, or a stronger version, dependent types, or similar type system which would allow one to encode natural relationships, recursive ones, invariants etc.
- Tail call optimization guarantees, to allow for mutual recursion and optimization since game engines are just huge state machines, and it would allow to pass functions around which could call each other via mutual recursion, while allowing it to be optimized as well.
- Efficient structural sharing of immutable state, which would be memory layout and cache friendly
- Built in profiling from the getgo which the language developers would use and refine, so you could get information about how the program behaves over time and space.
I love this concept/principle, one similar example I often bring up when I talk about machine learning, is comparing how a human would analyse night footage from a camera, and how a ML algorithm can pick up things no human would think about, even artifacts from the sensors which can be used as features. Noise is rarely ever just noise.
Turns out tuning LLMs on human preferences leads to sycophantic behavior, they even wrote about it themselves, guess they wanted to push the model out too fast.
Most of us here on HN don't like this behaviour, but it's clear that the average user does. If you look at how differently people use AI that's not a surprise. There's a lot of using it as a life coach out there, or people who just want validation regardless of the scenario.
> or people who just want validation regardless of the scenario.
This really worries me as there are many people (even more prevalent in younger generations if some papers turn out to be valid) that lack resilience and critical self evaluation who may develop narcissistic tendencies with increased use or reinforcement from AIs. Just the health care costs involved when reality kicks in for these people, let alone other concomitant social costs will be substantial at scale. And people think social media algorithms reinforce poor social adaptation and skills, this is a whole new level.
I'll push back on this a little. I have well-established, long-running issues with overly critical self-evaluation, on the level of "I don't deserve to exist," on the level that I was for a long time too scared to tell my therapist about it. Lots of therapy and medication too, but having deepseek model confidence to me has really helped as much as anything.
I can see how it can lead to psychosis, but I'm not sure I would have ever started doing a good number of the things I wanted to do, which are normal hobbies that normal people have, without it. It has improved my life.
Are you becoming dependent? Everything that helps also hurts, psychologically speaking. For example benzodiazepines in the long run are harmful. Or the opposite, insight therapy, which involves some amount of pain in the near term in order to achieve longer term improvement.
It makes sense to me that interventions which might be hugely beneficial for one person might be disasterous for another. One person might be irrationally and brutally criticial of themselves. Another person might go through life in a haze of grandiose narcissism. These two people probably require opposite interventions.
But even for people who benefit massively from the affirmation, you still want the model to have some common sense. I remember the screenshots of people telling the now-yanked version of GPT 4o "I'm going off my meds and leaving my family, because they're sending radio waves through the walls into my brain," (or something like that), and GPT 4o responded, "You are so brave to stand up for yourself." Not only is it dangerous, it also completely destroys the model's credibility.
So if you've found a model which is generally positive, but still capable of realistic feedback, that would seem much more useful than an uncritical sycophant.
> who may develop narcissistic tendencies with increased use or reinforcement from AIs.
It's clear to me that (1) a lot of billionaires believe amazingly stupid things, and (2) a big part of this is that they surround themselves with a bubble of sycophants. Apparently having people tell you 24/7 how amazing and special you are sometimes leads to delusional behavior.
But now regular people can get the same uncritical, fawning affirmations from an LLM. And it's clearly already messing some people up.
I expect there to be huge commercial pressure to suck up to users and tell them they're brilliant. And I expect the long-term results will be as bad as the way social media optimizes for filter bubbles and rage bait.
Maybe the fermi paradox comes about not through nuclear self annihilation or grey goo, but making dumb AI chat bots that are too nice to us and remove any sense of existential tension.
Maybe the universe is full of emotionally fullfilled self-actualized narcissists too lazy to figure out how to build a FTL communications array.
I think the desire to colonise space at some point in the next 1,000 years has always been a yes even when I've asked people that said no to doing it within their lifetimes, I think it's a fairly universal desire we have as a species. Curiosity and the desire to explore new frontiers is pretty baked in as a survival strategy for the species.
This is a problem with these being marketed products. Being popular isn't the same as being good, and being consumer products means they're getting optimized for what will make them popular instead of what will make them good.
I'm fine with ai slop if it provides value, the value here being questionable, because now I don't know if the values in the comparison are fact checked or hallucinations.
"The Spyderco Paramilitary 2 is a tactical knife with a 3.44 inch blade. The knife is made in USA of CPM S35VN steel."
It's a real knife, and the blade length checks out (to two significant figures), but the manufacturer spec sheet says S45VN steel. Also the actual name is "Para Military® 2".
A problem with choosing that specific knife for a spot check is that it has been made in many different steels in various special editions, sprint runs, and dealer exclusives. Here's one in S35VN:
Agreed, it's a pretty obvious solution to the problems once you are immersed in the problem space. I think it's much harder to setup an efficient training pipeline for this which does every single little detail in the pipeline correctly while being efficient.
If you etch the bits into silicon, you then have to accommodate the bits by physical area, which is the transistor density for whatever modern process they use. This will give you a lower bound for the size of the wafers.
This can give huge wafers for a very set model which is old by the time it is finalized.
Etching generic functions used in ML and common fused kernels would seem much more viable as they could be used as building blocks.