hatefulmoron

You're proposing a truism: if you don't get a good result, it's either because your query is bad or because the LLM isn't good enough to provide a good result.

Yes, that is how this works. I'm talking about the case where you're providing a good query and getting poor results. Claiming that this can be solved by more LLM conversations and ultrathink is cope.

I've read your questions a few times and I'm a bit perplexed. What kind of answers are you expecting me to give you here? Surely if you use Claude Code or other tools you'd know that the answers are so varying and situation specific it's not really possible for me to give you solid answers.

Distinction without a difference. I'm talking about its output being insufficient, whatever word you want to use for output.

I have a Claude max subscription. When I think of bad Claude code, I'm not thinking about unused variable definitions. I'm thinking about the times you turn on ultrathink, allow it to access tools and negotiate it's solution, and it still churns out an over complicated yet partially correct solution that breaks. I totally trust Claude to fix linting errors.

I don't really understand your point. It reads like you're saying "I like good code, it doesn't matter if it comes from a person or an LLM. If a person is good at using an LLM, it's fine." Sure, but the problem people have with LLMs is their _propensity_ to create slop in comparison to humans. Dismissing other people's observations as purely an emotional reaction just makes it seem like you haven't carefully thought about other people's experiences.

Calling things "slop" is just begging the question. The real differentiating factor is that, in the past, "human-generated slop" at least took effort to produce. Perhaps, in the process of producing it, the human notices what's happening and reconsiders (or even better, improves it such that it's no longer "slop".) Claude has no such inhibitions. So, when you look at a big bunch of code that you haven't read yet, are you more or less confident when you find out an LLM wrote it?

My mistake :-)

K et al. can look like that, but this example doesn't require any explanation to someone who has any familiarity.

I don't think we're saying different things. People who are distressed about their appearance are more likely to be depressed, and people who seek medicine and surgeries are probably more distressed still, and therefore more likely to be depressed, ..

There's probably a difference in degree, however. Alopecia Areata is much more uncommon, while regular male pattern baldness is very common.

There's also the fact that Alopecia Areata is actually more common in women, which I imagine exaggerates the distress compared to the more run of the mill MPB.

I realize you didn't mean to use a study on Alopecia Areata, but the difference in degree could be quite large.

It's also possible that people taking Finasteride might be a more potent selection of people that are distressed about hair loss, and are therefore more likely to exhibit depression, etc. As in, if people with androgenetic alopecia are more likely to be depressed, people who take finasteride may be a sampling of those people who are distressed enough to seek and maintain treatments.

I remember discussing with some coworkers a year(?) ago about autocomplete vs chat, and we were basically in agreement that autocomplete was the better feature of the two.

Since we've had Claude Code for a few months I think our opinions have shifted in the opposite direction. I believe my preference for autocomplete was driven by the weaknesses of Chat/Agent Mode + Claude Sonnet 3.5 at the time, rather than the strengths of autocomplete itself.

At this point, I write the code myself without any autocomplete. When I want the help, Claude Code is open in a terminal to lend a hand. As you mentioned, autocomplete has this weird effect where instead of considering the code, you're sort of subconsciously trying to figure out what the LLM is trying to tell you with its suggestions, which is usually a waste of time.

I often ask Claude to scan through the code first and then come back with questions related to the task. It sometimes comes back with useful questions, but most of the time it acts like a university student looking for participation marks from a tutorial; choosing questions to signal understanding rather than be helpful.

I can feel the ground rumbling as thousands approach to engage in a "name the trait" style debate..

It's often said that formal verification moves bugs from the implementation to the specification, and there's also the problem of proving equality between the formally proven specification and the implementation.

The first problem is just hard, the second can be approached by using a language like F* or Dafny that compiles to executable code. IIRC Amazon has had some success by using Lean for the specification and using fuzzing to test that the implementation matches.

Your prompting suggestion would certainly make them much better at this task, I would think.

I think it just points to the fact that LLMs have no "sense of self". They have no real knowledge or understanding of what they know or what they don't know. LLMs will not even reliably play the character of a machine assistant: run them long enough and they will play the character of a human being with a physical body[0]. All this points to the fact that "Claude the LLM" is just the mask that it will produce tokens using at first.

The "count the number of 'r's in strawberry" test seems to just be the easiest/fastest way to watch the mask slip. Just like that, they're mindlessly acting like a human.

[0]: https://www.anthropic.com/research/project-vend-1

It does say something that the models simultaneously:

a) "know" that they're not able to do it for the reason you've outlined (as in, you can ask about the limitations of LLMs for counting letters in words)

b) still blindly engage with the query and get the wrong answer, with no disclaimer or commentary.

If you asked me how many atoms there are in a chair, I wouldn't just give you a large natural number with no commentary.

Groq and Cerebras definitely have the t/s, but their hardware is tremendously expensive, even compared to the standard data center GPUs. Worth keeping in mind if we're talking about a $20 subscription.

They're using the term 'bullshit' as it is understood as a term of art, which doesn't imply lying. It's closer to creating a response without any regard for telling the truth. Bullshitting is often most effective when you happen to be telling the truth, although the bullshitter has no commitment to that.

> Isn't soreness basically guaranteed to some degree if you've done enough work to actually build strength?

Not really. If you're eating/sleeping well and training consistently it's completely normal to not feel soreness (that is, excluding the immediate discomfort that rapidly subsides). I can't speak for all forms of exercise, but certainly it's normal when lifting weights, even to failure.

That said, if you're just starting out you will notice a lot of soreness. Many people look back on the early DOMS and wish they could feel that sort of "positive feedback" again.

Is there any evidence that central planning on a much larger scale is drastically more efficient? We're talking about a whole country, after all. I take your point that companies themselves are usually centrally planned internally, but centrally planned economies haven't fared so well.

We have examples of countries with universal healthcare, long maternity leave, economic incentives to have children, and better social safety nets.

They have lower birth rates than the United States. Look at Sweden, Norway, and Finland.

I'm not saying these aren't good policies, but they aren't a fix for having more children.

But the point is that the illness of developed nations is having the symptoms treated by infusions of immigrants from countries that don't _yet_ have the same illness. If it's a matter of "as countries develop they have less children", you'd better hope other places of the world stay poor and uneducated.

> Also you’re interacting with the wrong kinda lady my friend.

When you sign a contract you have to be careful to consider how fucked you can be by the other side. It's not really the fault of that particular lady, you don't know how things will turn out.

Presumably either the countries being emigrating from must be at replacement rate, or themselves declining? Somebody has to be producing the next generation, somewhere..

"Just" is doing a lot of heavy lifting there. It definitely helps with getting data but actually training your model would be very capital intensive, ignoring the cost of paying for those outputs you're training on.

My concern isn't that it's messing up my architecture as I scream in protest from the other room, powerless to stop it. I agree with you and I think I'm being quite clear. Without relatively close guidance, it will paint you into a corner in terms of architecture. Guide it, direct it, whatever you want to call it.

> If you let it paint you into a corner, why are you doing so?

What do you mean? If it were as simple as not letting it do so, I would do as you suggest. I may as well stop letting it be incorrect in general. Lots of guidance helps avoid it.

> Maybe I just naturally do that, but to me there's not been much change in the level of guidance I need to give Claude Code or my own agent vs. what I'd give developers working for me.

Well yeah. You need to give it lots of guidance, like someone who works for you.

> the cost of telling it to just throw out the code it builds later and redo it with additional architectural guidance keeps dropping.

It's a moving target for sure. My confidence with this in more complex scenarios is much smaller.

It depends on what you're doing. If it's a simple task, or you're making something that won't grow into something larger, eyeballing the code and testing it is usually perfect. These types of tasks feel great with Claude Code.

If you're trying to build something larger, it's not good enough. Even with careful planning and spec building, Claude Code will still paint you into a corner when it comes to architecture. In my experience, it requires a lot of guidance to write code that can be built upon later.

The difference between the AI code and the open source libraries in this case is that you don't expect to be responsible for the third-party code later. Whether you or Claude ends up working on your code later, you'll need it to be in good shape. So, it's important to give Claude good guidance to build something that can be worked on later.

For OpenAI: short answer is no. From what I've seen, their biggest expense is training future models. If they stop that (putting aside the obvious downsides) they'd still be in the hole for a few billion dollars a year.

This is based on what I've read here: https://www.wheresyoured.at/openai-is-a-systemic-risk-to-the... (big AI bear, for what it's worth)

edit: Well, if they shed the other expenses that only really make sense when training future models (research, more data, fewer employees ..) they would be pretty close to break even.