It's not that the people don't genuinely believe what they're saying. It's that they've deluded themselves into thinking their ideological right is "for the kids".

There's always been Reefer Madness sorts of people. Sex, Drugs, Rock and Roll, Video Games, DnD, Rap Music, Homosexuality, and on and on. Today it's half woke mind virus and half DEI (for lack of a better term). Most of the people that spout this stuff genuinely believe they're fighting for the kids.

Bit of a disappointing article. I kept looking for a revelation or insight from the letters but it never came.

Don't rely on the author to hold your hand to an insight - what did you believe that this article challenged?

No I agree. The article seemed like it was going somewhere and then just sort of trailed off and never reached any sort of conclusion.

Autism is more broad-spectrum than just related to social processing. It's most visible in social processing because that's the cognitive area that humans have highly specialized in as a species, where expectations of performance are very high, and thus where deficiencies processing complex information in real-time are most visible. If we were birds, we'd probably think autism had something to do with flying. Instead, we are talking tribal apes, so when someone has the cognitive differences that lead to autism, we notice most strongly that they are having trouble being a normal talking tribal ape.

But the effects of autism are visible outside of social interaction too, with repetitive behaviors, intense focused interests, trouble with adapting to change, rigidity in lifestyle, etc.

It is possible though, to unify those things, and to see those other effects also as second or third order effects of the same underlying deficiencies that cause problems in social interaction. I believe, for instance, that our super power as neurotypicals is our ability to see, process, model and make sense, especially in real-time, of what's inside the minds of other people. In a way, we are wired to be comfortable with multiple worlds or perspectives around us, because we can see them, process them, and make sense of them. It makes sense to me that a person who is less good at this, will end up seeking a model of the world that is more rigid. If the worlds of other people around you seem chaotic to you, and uncomprehensible, then you will seek an environment and an understanding of the world that is more static, rigid or fixed. So, I think, at least on a conceptual level, it's possible to link the root causes of social problems to the root causes of the need for rigidity and stability.

That's a fair point.

Excellently put.

Autism is not the inability to process social cues. It seems more likely to be the difficulty to either filter or generalize sensory input.

The default mode network is so active that there is little bandwidth remaining for processing sensory input.

That's either schizophrenia, ADHD, or one of the many faces of anxiety.

Having a mother like that is probably the only legit disability in the article

Come now we're discussing an article where 4/10 students have accomodations. It can't be that hard to get.

Interesting that you find the alternative of 4/10 needing accommodations to be completely unrealistic.

There's neurodivergant and there's neurodivergant. I've definitely worked with oddballs and nerds and various atypical folks. But there's a massive gulf between them and someone with legit no questions about it autism.

I don't see how people don't see it. LLMs are a revolutionary technology and are for the first time since the iPhone are changing how we interact with computers. This isn't block chains. This is something we're going to use until something better replaces it.

I agree to some extent, but we’re also in a bubble. It seems completely obvious that huge revenue numbers aren’t around the corner, not enough to justify the spend.

There's a long list of things that have "replaced" humans all the way back to the ox drawn plow. It's not sane to be angry at any of those steps along the way. GenAI will likely not be any different.

it's plenty sane to be angry when the benefits of those technical innovations are not distributed equally.

It is absolutely sane to be angry at people's livelihoods being destroyed and most aspects of life being worsened just so a handful of multi-billionaires that already control society can become even richer.

The plough also made the rich richer, but in the long run the productivity gains it enabled drove improvements to common living standards.

We're talking miraculous level of improvement for a SOA LLM to run on a phone without crushing battery life this decade.

People are missing the forest for the trees here. Being the go to consumer Gen AI is a trillion+ dollar business. How many 10s of billions you waste on building unnecessary data centers is a rounding error. The important number is your odds of becoming that default provider in the minds of consumers.

It’s extremely easy to switch.

I used ChatGPT for every day stuff, but in my experience their responses got worse and I had to wait much longer to get them. I switched to Gemini and their answers were better and were much faster.

I don’t have any loyalty to Gemini though. If it gets slow or another provider gives better answers, I’ll change. They all have the same UI and they all work the same (from a user’s perspective).

There is no moat for consumer genAI. And did I mention I’m not paying for any of it?

It’s like quick commerce, sure it’s easy to get users by offering them something expensive off of VC money. The second they raise prices or offer degraded experience to make the service profitable, the users will leave for another alternative.

> The important number is your odds of becoming that default provider in the minds of consumers.

I haven't seen any evidence that any Gen AI provider will be able to build a moat that allows for this.

Some are better than others at certain things over certain time periods, but they are all relatively interchangeable for most practical uses and the small differences are becoming less pronounced, not more.

I use LLMs fairly frequently now and I just bounce around between them to stay within their free tiers. Short of some actual large breakthrough I never need to commit to one, and I can take advantage of their own massive spends and wait it out a couple of years until I'm running a local model self-hosted with a cloudflare tunnel if I need to access it on my phone.

And yes, most people won't do that, but there will be a lot of opportunity for cheap providers to offer that as a service with some data center spend, but nowhere near the massive amounts OpenAI, Google, Meta, et al are burning now.

The moat will be memory.

As a regular user, it becomes increasingly frustrating to have to remind each new chat “I’m working on this problem and here’s the relevant context”.

GenAI providers will solve this, and it will make the UX much, much smoother. Then they will make it very hard to export that memory/context.

If you’re using a free tier I assume you’re not using reasoning models extensively, so you wouldn’t necessarily see how big of a benefit this could be.

The moat is the chat history and the flywheel of user feedback improving the product.

Given how often smaller LLM companies train on the output of bigger LLM companies, it's not much of a moat.

LLMs complete text. Every query they answer is giving away the secret ingredient in the shape of tokens.

They all offer some "memory" cross chat now and they're all more annoying than helpful. Not really compelling. You can pretty easily export your chat if you want.

Just build more tulip fields/railways/websites no matter the cost - the golden age is inevitable! ;-)

> The important number is your odds of becoming that default provider in the minds of consumers.

Markets that have a default provider are basically outliers (desktop OS, mobile OS, search, social networks, etc).

All other industries don't have a single dominant supplier who is the default provider.

I am skeptical that this market is going to be one of the outliers.

All the other markets with a default provider basically rely on network effects to become and remain the default provider.

There is nothing here (in this market) that relies on network effects.

How big does an LLM need to be to support natural language queries with RAG?

My hot (maybe just warm these days) take is, the problem with voice assistants on phones is they have to be able to have reasonable responses to a long tail or users will learn not to use them, since the use cases aren’t discoverable and the primarily value is talking to it like a person.

So voice assistants backed by very large LLMs over the network are going to win even if we solve the (substantial) battery usage issue.

Why even bother with the text generation then? You could just make a phone call to an LLM with a TTS frontend. Like with directory enquiries back in the day. Which can be set up as easily as a BBS if you have a home server rack like Jeff Geerling makes youtube videos about.

1 800 CHAT GPT exists, I use it often enough to put it on speed dial on my flip phone

> Being the go to consumer Gen AI is a trillion+ dollar business.

I use LLM’s all day and a highly doubt this. I’d love to hear your argument for how this plays out.

The MacBook I use for work sucks and has weird issues when it wakes up from sleep. I've started having to restart my computer to fix them. I can't remember the last time I've had to do that.