German Bundeswehr uses AI to screen applicants for right-wing extremism | Hacker News Reader

German Bundeswehr uses AI to screen applicants for right-wing extremism

news.osna.fm

39 pointsAmazingTurtle60 comments

Open articleView on HN

Link to the actual article: https://archive.is/kt3Aa

How extreme do you have to be to get in?

Obvious AI generated article without citing any sources.

And no other osna.fm articles submitted to this site

seems legit https://derspiegel.substack.com/p/german-military-deploys-ai...

for context - Der Spiegel is one of the well respected publications in the region - https://en.wikipedia.org/wiki/Der_Spiegel

orig in DE - https://archive.is/kt3Aa

yeah the reason why i submitted the EN article and not original spiegel one is because HN isn't a german only platform and i found no english article on this, thought it was important news anyway

I think that's a violation of the AI Act.

At least in theory, no AI is making decisions on that - just adding a summary/categorization to the applicant dossier sitting in front of the HR. Now the bigger question is, what does a normal employee do with an AI recommendation? What are you as a software developer doing with the agent recommendations?

> If a story is spam or off-topic, flag it. Don't feed egregious comments by replying; flag them instead. If you flag, please don't also comment that you did.

Interesting, why is this spam or off topic? This is absolutely an AI topic that needs to be discussed. Not off-topic. Moderation? Hello?

The question could also be, is HN being targeted with bots, just like a Reddit, YouTube, Twitter, Facebook...

HN has far simpler controls compared sites like those.

there are a LOT of bots here, and if you understand what HN is, then it makes a lot of sense.

Do they check the actual posts and make sure they've got the right person? Or just go with the AI summary?

Every force has an equal and opposite force. The more you push, the more push-back you will get.

Human psychology isn't Newtonian physics, cultural norms are not a conserved quantity.

Curious to know which SOTA Frontier model did they use? Or an open weights model?

Why screen only the applicants and not the whole army?

ai prompt: "as a application reviewer for Deutsches Bundeswehr, search and identify applicants who have been rejected from Austrian art-schools." (I am not right-wing extreme)

Hopefully not musk's grok

Whos to say it's right or wrong? If I make a twitter profile impersonating a specific person and fill it with right-wing extremism will it get flagged by their process?

While technically possible, what are the chances to be targeted by such a defamation attack? We are not talking some high profile personality, just some Joe User looking for a job.

Pretty low, but it's damaging enough to worry about

Surely its just any postcode east of Berlin?

yeah you could basically go by city at this point

Systemic racism within the German system has been proven [1], so the Bundeswehr is right to screen for it. I have experienced racists at work, at government offices and in everyday situations. A system consisting of such people does not properly work. We need some Goodwill.

[1]: to name one: https://fgz-risc.de/forschung/inra-studie

Reply on news.ycombinator.com