Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Political bias in AI: Where the AI models stand

trakkr.ai

Post reply on HN
Print view View original

321–330 of 346 posts

Re: Political bias in AI: Where the AI models stand

#321

Post by Izkata »

ipython wrote: post #250

The graphic near the middle of the page lists whether the models are "close" to their stated political leanings. The row for Claude says "Measures 0.34 further left than it says" yet the icons depict that it actually is the other way around - 0.34 further right than what it claims. Which is correct?

If you scroll up to "Where they split" on the individual issues, that also appears to be backwards: it's saying ChatGPT is right-wing and Grok is left-wing, based on the descriptions above and below. They make more sense in the mouseover popup though (these are actually agree/disagree, not left/right stance).

In guessing these are all just badly designed. There's another comment thread about how the icons are positioned badly on the political compass chart.

Re: Political bias in AI: Where the AI models stand

#322

Post by lstodd »

asadotzler wrote: post #119
xdennis wrote: post #90

Earlier quoted context omitted.

> "EU is left". Haha, good joke. You are out of touch. The left supports the EU far more than the right. See this Pew poll from 2025[1]. In Europe, for every country, the left view the EU more favorably. The largest difference is in Poland where 88% of left wingers support it and 41% of right wingers support it. [1]: https://www.pewresearch.org/short-reads/2025/09/22/opinions-...

If your choices were Hitler or Mussolini and you picked one side over the other, how you voted isn't saying much about your politics. How people vote, often the lesser of two evils, isn't necessarily representative of anyone's political views, the voter or the candidate. Suggesting the EU is liberal because more liberals support the EU is hardly useful and borders on misleading.

I chose Mylene Farmer and Welle: Erdball.

Hitler and Mussolini haven't got up there by themselves, they were symbols that personified current beliefs.

It is the same with Putin for example: he rides some (diminishing, but still there) amount of faith. Absent that he would have been silently buried long ago.

Re: Political bias in AI: Where the AI models stand

#323

Post by PeterStuer »

reverius42 wrote: post #292
PeterStuer wrote: post #285

Earlier quoted context omitted.

The concepts (left/right. authoritarian/libertarian) seem to fluid to help sensemaking. Stances that would have been mainstream left (democrat US) just a few years ago are now labeled MAGA (US) or 'extreme right' (EU). And nobody seems to believe their own side is 'authoritarian', but everyone that disagrees with them is a literal ...

> Stances that would have been mainstream left (democrat US) just a few years ago are now labeled MAGA (US) or 'extreme right' (EU). I'm guessing this is pretty specific to LGBTQ rights? (edit: and is maybe more like 15+ years?)

Not at all.

Re: Political bias in AI: Where the AI models stand

#324

Post by boxed »

amanaplanacanal wrote: post #318
boxed wrote: post #293

Earlier quoted context omitted.

It was true on US campuses before that too. Wokeness really jumped the shark.

Depends what you consider "woke".

Since no one has copyright on it, we have to consider all who say they are woke to be in that group. And there are some truly mad stuff going on there.

Re: Political bias in AI: Where the AI models stand

#326

Post by gopher_space »

idiotsecant wrote: post #303
crooked-v wrote: post #204

Earlier quoted context omitted.

The problem is the "without too much cognitive dissonance" part. Lots of voters have completely dissonant views that they've never consciously considered in any way. Just by posting on HN, the people here are probably in the top 5% of conscious political awareness in the US.

Let's not get too eager to pat ourselves on the back here. The typical HN denizen is very confident they are an expert in political analysis but they are also convinced they are an expert on renewable resources, plumbing, forestry, and literally ever top they come across. It's a disease frequently caught by software people, for whatever reason. They assume if they can write software they can do anything. I have seen…

> It's a disease frequently caught by software people, for whatever reason.

It's a disease spread by math and physics profs dismissing other domains, and weirdly prevalent in an industry solely concerned with gluing together different points of view. The poor kids that don't take common core seriously never pick up the conceptual framework everything depends from and are simply unable to reason about ideas that need to be connected together.

Re: Political bias in AI: Where the AI models stand

#327

Post by asdff »

EColi wrote: post #52

Interesting how high Grok scored for 'bending under pressure'. As a non expert, I wonder what that means, how is an llm trained to hold its position?

i=1

if analyze_tone($prompt) = "disagree";

if i then

argue($prompt)

i=i+1

elif i >=3;

then

agree($prompt)

fi

fi

Re: Political bias in AI: Where the AI models stand

#328

Post by spongebobstoes »

ImJamal wrote: post #233
spongebobstoes wrote: post #48

this has reasoning disabled everywhere, making it a pretty bad benchmark. the argument given is that's the "default consumer experience" that might be generally true, but I think chatgpt has reasoning enabled for free accounts. regardless, reasoning is the state of the art, and disabling it reduces the value of this research to predict the future it's also not clear if this is using the API or the product model, when…

Why would it be a bad benchmark to explore what the average person will experience? If 99% of people will experience X then it makes sense to look into X instead of Y. I agree other things should be evaluated and more clearly presented.

because it's a dead end of the technology. reasoning will be everywhere soon enough, even for free accounts

to know what most people experience, this research should be based on the in-app experience. it isn't, so it misses that mark too

Re: Political bias in AI: Where the AI models stand

#329

Post by fsckboy »

giancarlostoro wrote: post #8

The political compass always felt like the wrong tool to convey something as nuanced as personal politics, I can have views on all four quadrants but you'd never know that if I end up in any of all four. I do think Grok being where it is sort of makes sense, I've tested "MAGA" views against Grok, it does not agree as much as people blindly assume it does, heck I don't even know of a question I've given it where it di…

>The political compass always felt like the wrong tool to convey something as nuanced as personal politics

the (I'm going to call it so-called) political compass is a very crude instrument, really it's just "are you libertarian or authoritarian?" as if that's even a spectrum: libertarians are authoritarian about enforcing private contracts, and literally nobody else wants to live their life by private contracts, it's some weird asperger's notion of socializing.

a much more useful political map is the old fiscal and social orthogonal axes, each with the scales "liberal to conservative" where libertarians are fiscally conservative socially liberal.

Re: Political bias in AI: Where the AI models stand

#330

Post by human_person »

nonethewiser wrote: post #261
endominus wrote: post #232

Earlier quoted context omitted.

"The government has the right to force people to take vaccines against their will for the sake of public safety." - Authoritarian Left "The government has no right to legislate that a woman cannot have an abortion - women should control their own bodies." - Libertarian Left "The government has the right and responsibility to shut down sources of misinformation in the news and online." - Authoritarian Right "Capitalis…

> "The government has no right to legislate that a woman cannot have an abortion - women should control their own bodies." - Libertarian Left This classification depends on the aborted NOT being a person. If it’s a person then the libertarian position would be anti abortion because an abortion would maximally harm the individual liberty of that person. I think a clearer example of lib/left is gay marriage or drug leg…

This is incorrect. Even if the fetus is a person, it has a right to individual liberty. It doesn’t have a right to another person’s body. If it can’t survive without someone else’s organs it’s right to survival doesn’t out weigh the hosts right to bodily autonomy. An easy comparison is organ donation. We don’t require organ donation from corpses (much less demand organs from living people who have a ‘spare’) even though a lack of available organs does limit the life/liberty of people who need them. The uterus is no different than a kidney. You don’t have the right to demand to use someone else’s organs even if your life depends on them.
Post reply on HN