Injecting bias into an already biased model doesn’t make decision smarter, it just makes them faster.

I think this continued anthropomorphism "Have you tried asking about..." is a real problem.

I get it. It quacks like a duck, so seems like if you feed it peas it should get bigger ". But it's not a duck.

There's a distinction between "I need to tell my LLM friend what I want" and "I need to adjust the context for my statistical LLM tool and provide guardrails in the form of linting etc".

It's not that adding prose description doesn't shift the context - but it assume a wrong model about what is going on, that I think is ultimately limiting.

The LLM doesn't really have that kind of agency.

It’s not a thank you letter. It’s AI slop.

LLMs are plateauing, and you’re in denial.

Show me one metric they are plateauing on.

Like genital herpes, I am stuck with it. I might as well learn to like it. And eventually convert others across to the cause.

Remember to disclose, I'm not a lawyer, but pretty sure it's criminal to spread without informing and getting consent.

Your employer needs to know before you infect their codebase with react.

Once in a blue moon is fine but at that frequency? Sounds like you’re just short circuiting your brain and enjoying the sprays while they last.

> Once in a blue moon is fine but at that frequency? Sounds like you’re just short circuiting your brain and enjoying the sprays while they last.

Actually, the first time I used LSD that was partly the goal; I was trying to find some way to break my brain or otherwise manufacture some brand of insanity. For me though, it turned out not to be particularly capable of that on its own, or at least it doesn't seem to trigger that to happen. It did do something else that I ended up still liking, and that something else is what I use it for nowadays.

Ones that are offended by a lighthearted joke. Sounds like a plus

What are you even trying to say with that? The troll got what they wanted, attention.

We are not stochastic parrots. Old components of our brain help “ground” our thoughts and allow things like doubt or a gut feeling to develop which means we can question ourselves in ways an LLM cannot.

There’s a fine line between ‘open’ and ‘perverted’.

Perverted is just some other person being judgmental.

That line is shared with the one of "none of your business"

Yep. That's why I'm openly talking about moderation and why I'm very much in favor of an NSFW filter.