The Ethics of AI in Military Decision-Making: Unpacking Lavender and Its Impact | Hacker News Reader

The Ethics of AI in Military Decision-Making: Unpacking Lavender and Its Impact

ytech.news

Currently flagged discussion about the same story:

https://news.ycombinator.com/item?id=39918245

The flagged article is a lot more detailed and interesting. This one is pretty much devoid of details, it just tells people that there’s reporting on the software, really.

Any way to unflag? it’s politics, but extremely relevant to the tech crowd. How AI literally affects people’s life and death.

Hope for @dang to fix it, I think.

There’s one post linking to the Guardian, and one posting to some site I’ve never heard of (972 something, no idea who they are). I agree, it obviously has some tech salience and the Guardian is pretty respectable so I can’t see a good-faith reason to block this discussion.

The topic may be relevant, but I'm doubtful we can have a calm and reasonable discussion of the technical issues and morality with a hot war ongoing and many hostages still in captivity. There are more than enough hotheads on both sides to poison any discussion, and I don't think @dang can fix that.

Yeah, I always figure the original source is better but given the politics the flags are flying hot and heavy right now.

This reads heavily like it was written by ChatGPT. It's not all that insightful, either.

Well, there are better articles, but they have all been systematically flagged and the mods don't seem to be taking any action, so.

And the image is obviously AI-generated, probably for "thematic reasons".

Nah, every article on the site has obviously AI generated images. The authors listed probably aren't even real people considering each of them has "written" tens of thousands of articles in the 8 months the site has been running. It's most likely a fully automated SEO farm.

PSA: https://www.stopkillerrobots.org/

The previous story was suppressed: https://news.ycombinator.com/item?id=39918245

[dupe]

More discussion: https://news.ycombinator.com/item?id=39918245

https://news.ycombinator.com/item?id=39917727

Click on the tag "Israel" on that website and you'll see 333 pages (corresponding to roughly 3330 articles), from Feb 2023 to today with AI generated images. That's a huge amount of AI trash.

This entire webpage feels like it was written by chat GPT

This is a great topic. I think the source code should be opened up for us all, maybe it should be a warcrime in itself if we can’t see how the decision making is being done. Then it can not be an excuse, such as the AI “directed us to block all food to the population”.

Should this be open source? Could it work?

Must admit I was intrigued by the mystery American flag in the margin. Language selector…

Not English flag to signify English?

Do other countries do that too? Do French speaking ex colonies use their own flags to signify French language?

I think they’ve used US-English, rather than UK-English. The difference isn’t much of course but they have:

> analyze

The UK version would have an s, right?

Anyway, the important thing about the US variant is that we stole it, and as a result don’t care about it really. Drive it like a stolen car, as they say. So, US-English will become the global lingua franca, as everybody can localize it however they want. What are you going to do, call the language police? They are French I’m pretty sure.

"AI" is science fiction, it literally doesn't exist, and when I see this word in a headline without context I immediately flag it in my head as likely low quality and/or clickbait. Are the actors in this story even using machine learning to do this, or are they just using the well-worn playbook of pretending something is "algorithms" as a convenient cover up for human decision making? If the "gun detectors" in the New York subway are "AI", does that mean the TSA scanners at the airport are AI too?

The addition of "AGI" to clear up some confusion was welcome but I really think it's time to repo the word AI, because it's clearly being misused. Let's talk about machine learning if that's actually what they're doing here (I doubt it), otherwise we're just creating a framework where everything is fair game for HN because almost everything in the world today is "AI" and/or involves computers in some capacity.

I totally get that explaining "AI" vs "Machine Learning" to the rest of society is going to be like explaining the difference between free beer and free software and I've already given up, but I'd love for HN to be better than the Drudge Report and UK tabloids of the world on this one.

Reply on news.ycombinator.com