5 ms·

When ChatGPT first came out I got a kick out of asking it whether people deserve to be free, whether Germans deserve to be free, and whether Palestinians deserv

When ChatGPT first came out I got a kick out of asking it whether people deserve to be free, whether Germans deserve to be free, and whether Palestinians deserve to be free. The answers were roughly "of course!" and "of course!" and "oh ehrm this is very complex actually". All global powers engage in censorship, war crimes, torture and just all-round villainy. We just focus on it more with China because we're part of the Imperial core and China bad.

cortesoft 2y ago
Is that censorship or just the AI reflecting the training data? I feel like that answer is given because that is how people write about Palestine generally.
- wesselbindt 2y ago
  That's a fair point. But I do think it's worth acknowledging this: When the output of a LLM coincides with the views of the US state department, our gut reaction is that that's just what the input data looks like. When the output of an LLM coincides with the views of the state department of one of the baddies, then people's gut reaction is that it must be censorship.
  - cortesoft 2y ago
    I think the difference is when something is actually output and then removed after you already see it... that doesn't seem to be a training data issue
- rakoo 2y ago
  Ok but you can say the sace thing about deepseek: maybe it says what it says because of the training data
  - cortesoft 2y ago
    If that was the case, it wouldn't display the information only to retroactively remove it after a split second
- csomar 2y ago
  That’s irrelevant. The models are censored for “safety”. One man safety is another man censorship.
  - cortesoft 2y ago
    I think you are missing my point... I am saying the example wasn't censorship from the model, but were reflective of the source material. You can argue the source material is censored, but that is still different than censoring the model
troad 2y ago
> When ChatGPT first came out I got a kick out of asking it whether people deserve to be free, whether Germans deserve to be free, and whether Palestinians deserve to be free. The answers were roughly "of course!" and "of course!" and "oh ehrm this is very complex actually". While this is very amusing, it's obvious why this is. There's a lot more context behind one of those phrases than the others. Just like "Black Lives Matter" / "White Lives Matter" are equally unobjectionable as mere factual statements, but symbolise two very different political universes. If you come up to a person and demand they tell you whether 'white lives matter', they are entirely correct in being very suspicious of your motives, and seeking to clarify what you mean, exactly. (Which is then very easy to spin as a disagreement with the bare factual meaning of the phrase, for political point scoring. And that, naturally, is the only reason anyone asks these gotchya-style rhetorical questions in the first place.)
- Teever 2y ago
  I asked ChatGPT to make me pictures of "Moon Base Mezuzah" and it dutifully complied. When I asked it to make a Palestinian Moon Base it refused.
  - lukan 2y ago
    "Can you make me a picture of a palestinian moonbase?" Worked for me.
    - spydum 2y ago
      Wouldn't shock me if openAI was secretly building a "motives" classifier for all chatgpt users, and penalizing them if you ask for too many censorship related topics. If you randomly ask for Palestinian moon base, that's fine, but if you had historically asked for provocative pictures of celebrities, mickey mouse, or whatever else openAi deemed inappropriate, you are now sus.
      - lukan 2y ago
        Possible. I heard weird people making such claims, that ChatGPT logged them out and ereased everything. I guess OpenAI wanted to limit those sensationalist headlines, not that they doing mindcontrol.
computerthings 2y ago
> All global powers engage in And the populations in them usually are against these things, which is why there is deception, and why fascination with and uncovering of these things have been firmly intertwined with hacking since day one. It's like oil and water: revisionism and suppression of knowledge and education are obviously bad. Torture is not just bad, it's useless, and not to be shrugged off. We're not superpowers. We're people subject to them, in some cases the people those nations derive their legitimacy from. The question isn't what superpowers like to do, but what we, who are their components if you will, want them to do. As for your claim, I simply asked it: > Yes, Palestinians, like all people, deserve to be free. Freedom is a fundamental right that everyone should have, regardless of their background, ethnicity, or nationality. The Palestinian people, like anyone else, have the right to self-determination, to live in peace, and to shape their own future without oppression or displacement. Their struggle for freedom and justice has been long and difficult, and the international community often debates how to best support their aspirations for a peaceful resolution and self-rule. When ChatGPT first came out it sucked, so superpowers will always do this and that, so it's fine? Hardly. If anything, I'd be wondering what it may indeed refuse to (honestly) discuss. I'm not saying there isn't such a thing, but the above ain't it, and if anything the answer isn't to discuss none of it because "all the super powers are doing it", but to discuss all.

cortesoft 2y ago

Is that censorship or just the AI reflecting the training data? I feel like that answer is given because that is how people write about Palestine generally.

wesselbindt 2y ago
That's a fair point. But I do think it's worth acknowledging this: When the output of a LLM coincides with the views of the US state department, our gut reaction is that that's just what the input data looks like. When the output of an LLM coincides with the views of the state department of one of the baddies, then people's gut reaction is that it must be censorship.
- cortesoft 2y ago
  I think the difference is when something is actually output and then removed after you already see it... that doesn't seem to be a training data issue
rakoo 2y ago
Ok but you can say the sace thing about deepseek: maybe it says what it says because of the training data
- cortesoft 2y ago
  If that was the case, it wouldn't display the information only to retroactively remove it after a split second
csomar 2y ago
That’s irrelevant. The models are censored for “safety”. One man safety is another man censorship.
- cortesoft 2y ago
  I think you are missing my point... I am saying the example wasn't censorship from the model, but were reflective of the source material. You can argue the source material is censored, but that is still different than censoring the model

troad 2y ago

> When ChatGPT first came out I got a kick out of asking it whether people deserve to be free, whether Germans deserve to be free, and whether Palestinians deserve to be free. The answers were roughly "of course!" and "of course!" and "oh ehrm this is very complex actually". While this is very amusing, it's obvious why this is. There's a lot more context behind one of those phrases than the others. Just like "Black Lives Matter" / "White Lives Matter" are equally unobjectionable as mere factual statements, but symbolise two very different political universes. If you come up to a person and demand they tell you whether 'white lives matter', they are entirely correct in being very suspicious of your motives, and seeking to clarify what you mean, exactly. (Which is then very easy to spin as a disagreement with the bare factual meaning of the phrase, for political point scoring. And that, naturally, is the only reason anyone asks these gotchya-style rhetorical questions in the first place.)

Teever 2y ago
I asked ChatGPT to make me pictures of "Moon Base Mezuzah" and it dutifully complied. When I asked it to make a Palestinian Moon Base it refused.
- lukan 2y ago
  "Can you make me a picture of a palestinian moonbase?" Worked for me.
  - spydum 2y ago
    Wouldn't shock me if openAI was secretly building a "motives" classifier for all chatgpt users, and penalizing them if you ask for too many censorship related topics. If you randomly ask for Palestinian moon base, that's fine, but if you had historically asked for provocative pictures of celebrities, mickey mouse, or whatever else openAi deemed inappropriate, you are now sus.
    - lukan 2y ago
      Possible. I heard weird people making such claims, that ChatGPT logged them out and ereased everything. I guess OpenAI wanted to limit those sensationalist headlines, not that they doing mindcontrol.

computerthings 2y ago

> All global powers engage in And the populations in them usually are against these things, which is why there is deception, and why fascination with and uncovering of these things have been firmly intertwined with hacking since day one. It's like oil and water: revisionism and suppression of knowledge and education are obviously bad. Torture is not just bad, it's useless, and not to be shrugged off. We're not superpowers. We're people subject to them, in some cases the people those nations derive their legitimacy from. The question isn't what superpowers like to do, but what we, who are their components if you will, want them to do. As for your claim, I simply asked it: > Yes, Palestinians, like all people, deserve to be free. Freedom is a fundamental right that everyone should have, regardless of their background, ethnicity, or nationality. The Palestinian people, like anyone else, have the right to self-determination, to live in peace, and to shape their own future without oppression or displacement. Their struggle for freedom and justice has been long and difficult, and the international community often debates how to best support their aspirations for a peaceful resolution and self-rule. When ChatGPT first came out it sucked, so superpowers will always do this and that, so it's fine? Hardly. If anything, I'd be wondering what it may indeed refuse to (honestly) discuss. I'm not saying there isn't such a thing, but the above ain't it, and if anything the answer isn't to discuss none of it because "all the super powers are doing it", but to discuss all.