sumeno

10 ms·

151.
by sumeno 6mo ago
LLMs don't understand what they are doing, they can't explain it to you, it's just creating a reasonable sounding response
152.
by sumeno 6mo ago
Faster doesn't always mean better. I've "learned" things from LLM really fast, but I don't retain the information the same way as if I had taken my time to really work through it
153.
by sumeno 6mo ago
That's what the program he just took was supposed to be for, learning not output. You've just reinvented the article from first principles, congrats
154.
by sumeno 6mo ago
He tried and messed that up too
155.
by sumeno 6mo ago
So many comment sections on these vibe coded Show HNs are just full of obvious bots talking to each other in the most generic boring way.
156.
by sumeno 6mo ago
At least in the novel example the author had the decency to write what they're asking you to read. These are more like sending someone who didn't ask you a question a LMGTFY link they didn't ask for and expecting them to read
157.
by sumeno 6mo ago
> if it works it works If it works in one case that doesn't mean it works consistently or well in the general case I've made lots of things with Claude Code that just work... until I do things in a slightly different order and
158.
by sumeno 7mo ago
> It's useful context unless you've gone over the generated code and understand it and it is the same quality as if you wrote it yourself If this is not the case you should not be sending it to public repos for review at all. I
159.
by sumeno 7mo ago
No real reason to do that, they say Claude Code is written by Claude, which means it has no copyright. Just use the code directly
160.
by sumeno 7mo ago
> my brain has not deteriorated now that i have offloaded that to my phone. Is there empirical evidence that you haven't? You wouldn't necessarily be the best judge
161.
by sumeno 7mo ago
If you were Claude you would have no emotions or thoughts about a prompt one way or another
162.
by sumeno 7mo ago
How many dead from the US's proxy wars?
163.
by sumeno 7mo ago
There is no way to ensure that the AI doesn't guess. You can do all the prompting and RAG you want but sometimes it's just going to make shit up and ignore instructions
164.
by sumeno 7mo ago
Mugging victims didn't make a choice that endangered a bunch of other people that resulted in them getting mugged. Interlock devices are not given to random people for no reason.
165.
by sumeno 7mo ago
Considering you can't have chat bots be honest even if you don't make them marketing tools I suspect I know which way companies are going to go
166.
by sumeno 7mo ago
The only way the people who are only in it for the money leave the industry is if the money gets worse. If the money stays the same why would they leave
167.
by sumeno 7mo ago
No, only caffeinated > decaffeinated coffee intake was not associated with lower dementia risk or better cognitive performance
168.
by sumeno 7mo ago
The studies compared people from the same occupation, so no, that is not likely the reason
169.
by sumeno 7mo ago
> Since when did we restrict people's ability to do things? At least 4000 years ago, but that's just the earliest we have evidence for https://en.wikipedia.org/wiki/Code_of_Ur-Nammu
170.
by sumeno 7mo ago
LLMs are not people. They do not learn the way people do. Even if they did, if someone memorized copywritten code and then typed it back out that would still be a copywrite violation
171.
by sumeno 7mo ago
It's controlled by a guy who spends all day retweeting white supremacists and lying about his companies. Why should anyone who isn't a white supremacist use it?
172.
by sumeno 7mo ago
What % of that is owned by employees that aren't named Elon Musk?
173.
by sumeno 7mo ago
Are you doing a full search of every GPL licensed repository every time you use an LLM to ensure that it isn't giving you GPL licensed code? That doesn't seem reasonable
174.
by sumeno 7mo ago
"in accordance with its license" is the key part that's missing with LLMs. The licenses are completely ignored.
175.
by sumeno 7mo ago
> He says it's a gift, and if people do whatever, he doesn't care; he already gave it away. That's his choice and I assume he licensed his code accordingly. That doesn't mean that the choices of others who used differ
176.
by sumeno 7mo ago
How would you know whether or not it is AI slop, you haven't reviewed it
177.
by sumeno 7mo ago
You shouldn't ask other people for feedback before you have done the bare minimum of reading your "own" book
178.
by sumeno 7mo ago
Ahh, "you're holding it wrong" The classics never go out of style
179.
by sumeno 7mo ago
This has basically been my experience since Sonnet 3.5. I've been working on a personal project on and off with various models and things since then and the biggest difference between then and now is that it will do larger chunks of wo
180.
by sumeno 7mo ago
There are no good cops

More ›