@elmiko For now. As I said to @davecb, it will just turn into another spam-type arms race. A technological advance in detecting slop will just lead to technological advances in trying to make slop undetectable. And on and on. We've seen this story before with spam, adware, phishing, etc.

The problem isn't LLMs. It's how people are going to use them. If the evidence is that people will use them irresponsibly, then it's irresponsible to keep advancing the technology and promoting it, IMO.

Perhaps -- maybe -- fans of LLMs will see the maintainers of open-source projects that they care about complaining about the slop deluge and stop throwing slop over the walls, and tell their fellow fans that they should cool it. I'm not optimistic on that point, though, because all evidence so far points to LLM-user culture being one of disrespect for people outside the LLM bubble and flagrant disregard for the concept of consent.

The attitude seems to be "we like this, and so should you, and we're not going to respect your wishes. Now shut up and consume the slop we serve you." I'd love to be proven wrong on this.

@jzb @davecb i agree, it's a human problem not a technical problem.

> The attitude seems to be "we like this, and so should you, and we're not going to respect your wishes. Now shut up and consume the slop we serve you." I'd love to be proven wrong on this.

i also would love for you to be wrong, but i fear not. humans are not good at self-regulating nor self-reflecting.