As an example, imagine a really small community maintaining a small site/wiki/cms/forum that has the ultimate goal of promoting human relationships around a common interest (let's say retro computing as an example, but it could be anything). There are many many many such communities on the internet.
It's very improbable you will specifically instruct your LLM to access their site, it's way more probable it will happen without you even knowing, as a result of you doing some /deep-research or something. And not only that, but your LLM will probably spawn a ton of agents to gather as much information as possible in as little time as possible. A torrent of requests will go at this community's site, effectively killing it. They are a small community, they use their spare time and money to maintain something to serve them, they don't have the resources to serve your LLM and until you showed up they probably never even had to think about Cloudflare. They are certainly not against you getting the content, but they don't want you causing them issues either and you just did.
End result? Your LLM (effectively you) DoSed a small community's site. You caused harm. Could you have caused similar harm if you were doing it on your own? Sure. But you would have done it on purpose, not accidentally while instructing your LLM to do something else.
This isn't a made up story, it has happened already more than 1 times.
So the question is, now that you know your LLM can cause harm without you even knowing it, how does this change your stance?
We can stop putting information in the open in any form, as well.
FAIR: Findable, Accessible, Interoperable, Reusable.
Considering that, shall we paint the boards black or lock the doors to community spaces?
Also, while I assume this is not your first account here, these are worthy of reminding:
> Be kind. Don't be snarky. Converse curiously; don't cross-examine. Edit out swipes.
> Throwaway accounts are ok for sensitive information, but please don't create accounts routinely. HN is a community—users should have an identity that others can relate to.
For more, please refer to https://news.ycombinator.com/newsguidelines.html
Is defeating captchas unethical? I don’t think so. Not on its own.
Is scraping unethical? I don’t think so. Not on its own.
Are there tons of uses for both of those that are sketchy or outright wrong? Yeah, absolutely.
1. People like other people.
2. Businesses need to sell product
The internet mixes those people, and an Agent basically pushes the signal to noise ratio that businesses have relied on the intenet to basically zero.
If the internet just allowed indescriminate traffic, neither #1 nor #2 survives, and while it's interesting to think the value is, it certainly isn't anything we understand.
Maybe you think the value will still exist, but some how agents will replace all value with equivelents. That's an argument, but I don't think it will.
Sure, Nazi, Hitler or Stalin all considered themselves to be the good guys. Kushner, Trump, etc are also acting within their ethical system (if I am getting money or power or fame it is ok to do it). Just about the only exception is Thiel who openly frames himself as evil, but is proud of it.
That does not mean we cant criticize their crappy actions or "ethical frameworks".
And yes, all the above are making deliberate decisions to be unethical assholes.