Conservatives think ChatGPT has gone 'woke'

Ok, so I guess Hacker News has decided that data can only be 100% biased or 0% unbiased, but nothing in-between.

Yes, almost all data is biased... of course.

Some data is 100% biased. Some data is 1% biased.

How about we try to collect data and then weigh it such that what we feed to the AI during training is as unbiased as possible, given a certain amount of effort?

You know that you can actually influence what data you feed to the AI, right? Or how much the training takes some data into account vs some other data, I guess.

You know that you can create a metric for measuring bias, right?

You know that even if you are not capable of being 100% unbiased, you can work towards that goal, right?

You know that there are plenty of smart people who can come up with ideas for eliminating (or mitigating) sources of errors when measuring bias, right?

I hope you begin to see the solution at hand.

> The approach you describe has the problem that it's asking majority people about the experiences of minority folks

No, my approach is asking all types of people about the experience of minority folks, including those minority folks (we are all minority folks in some aspect, even if this aspect is uninteresting).

> for instance, if you ask a statistically significant sample of the population about what it is like to be a trans man, you are going to (...) be asking a bunch of people who have no idea what it is like.

Then those people can answer that they don't know what it's like to be trans.

If somebody comes up to me and asks me: "what is it like to be trans?". My answer would obviously be: "how the hell should I know? I'm not trans".

But trans people can answer what it's like to be trans.

> And it gets worse. For instance, trans men have a totally different experience in rural vs coastal America vs Europe vs Africa. To get an AI who can speak confidently on what it is like to be trans male in those places will require even more interviews.

Yes, you can only spend a limited amount of effort towards the goal of being unbiased. The goal is to be as unbiased as possible given that limited amount of effort.

It's still better to make X amount of effort to be unbiased than zero effort.

This is also something that can be improved over time, as better ideas and methods become available regarding how to measure and decrease bias.

Perhaps even an AI can be used to detect these biases and reduce them as best possible.

> Now, you could say, "My AI simply will not speak about the experience of gay Black men, and the challenges/perspectives from that community", but then you've introduced a bias.

Or perhaps the AI can simply answer based on the information it was trained on, making a best guess as to what that would be like, taking into account all the data that was available to it and how that data was weighed to be as unbiased as possible.

> You could say, "Well, we'll go out and interview people from every set then, make sure we're covering everyone!"

No, I think you are making a significant mistake in this reasoning. There is no "every set". There is only one set. And that is the set of all people.

> But where then do you stop sampling? Each additional modifier adds exponential complexity -- gay Black men from New Orleans will have a different experience from gay Black men from Lagos.

What modifier? There is no modifier. "SELECT RANDOM(x%) FROM TABLE all_people" (or whatever the imaginary SQL syntax would be) :)

since you're not seeing it, here's the pattern:

first, the New Current Thing drops—out of nowhere. overnight, something that was a complete non-issue mere hours before, is suddenly a super important issue that everyone needs to have an opinion on. facts and figures are presented with minimal if any context, academic rigor, or peer review. half of the population believes it all, 100%, at face value, because these are Scientific Facts and Figures. there is zero admission of having just believed something completely different only hours earlier, possibly for their entire lives up until the day of the New Current Thing dropping.

the other half of the population does not believe these things, continues to have the same opinions about the New Current Thing as they did the day before, and finds joy in being as obstinate about it as possible on social media.

in mere weeks, if that, this New Current Thing will be completely forgotten—there will be no change of public policy, but politicians "on each side" may return to "their side's" take on the matter in future debates. rather, we will have moved on to the Next New Current Thing, which will follow almost the exact same pattern. (though, the "sides" may be reversed, depending on the topic at hand.)

no positive change to society is achieved as a result. the only change is that people now have yet another reason to dislike each other, yet another insult-arrow in their get-mad-on-social-media-quiver. the chasm between the two common broad perceptions of reality widens.

take note the next time a New Current Thing drops, and see how closely it dropping and the discourse surrounding it hew to this general heuristic. if you allow yourself to examine these things dispassionately and remove yourself from the resulting emotionally-charged discourse, you might notice that this sort of thing happens more frequently than you'd think. you'll start to become shocked at what people are willing to immediately believe and internalize as fact, wholesale, with merely the slightest possible nudging—and how the other side is content with merely hurling sarcastic insults right back at the other side, completely unaware that their side of the public discourse's role is also fully intentional, entirely planned for.

it's all about reducing the signal-to-noise ratio and deepening interpersonal division.

Most political conflict is based in subjective values where there is no right or wrong, in the objective sense. It is more about what people want, or more cynically, don't want. In most cases, you can't prove that someone doesn't want what they want, and vice versa.

I think the most interesting and legitimate conversations in this space are those where people genuinely try to understand what others want, and seek out areas where they agree and have common ground.

This is difficult

I can get ChatGPT to mock men but not women.

How is that appropriate?