Dr. Jordan B. Peterson

Yes, but they're the ones that I also think that they're sort of oblivious nice.

Yes, but they're the ones that I also think that they're sort of oblivious nice.

I think that's true And we that is what we found we found in the study. I got a research idea. Yes. Yes It's relevant to this well with regards to these questionnaires. It's something that I wanted to do You know the large language models track statistical probability so you can take those left-wing Questionnaire sets and you can ask chat GPT. Here's an item or here's three items generate 30 more

I think that's true And we that is what we found we found in the study. I got a research idea. Yes. Yes It's relevant to this well with regards to these questionnaires. It's something that I wanted to do You know the large language models track statistical probability so you can take those left-wing Questionnaire sets and you can ask chat GPT. Here's an item or here's three items generate 30 more

I think that's true And we that is what we found we found in the study. I got a research idea. Yes. Yes It's relevant to this well with regards to these questionnaires. It's something that I wanted to do You know the large language models track statistical probability so you can take those left-wing Questionnaire sets and you can ask chat GPT. Here's an item or here's three items generate 30 more

and it doesn't. Yeah, yeah, yeah. Right, so if you wanted to improve the statistical reliability of the measures, so you can imagine, take the measures that already exist, put them in clumps of three in ChatGPT, have it expanded out to like 300 items, administer it to a thousand people and distill it. Because the thing- That's a great idea.

and it doesn't. Yeah, yeah, yeah. Right, so if you wanted to improve the statistical reliability of the measures, so you can imagine, take the measures that already exist, put them in clumps of three in ChatGPT, have it expanded out to like 300 items, administer it to a thousand people and distill it. Because the thing- That's a great idea.

and it doesn't. Yeah, yeah, yeah. Right, so if you wanted to improve the statistical reliability of the measures, so you can imagine, take the measures that already exist, put them in clumps of three in ChatGPT, have it expanded out to like 300 items, administer it to a thousand people and distill it. Because the thing- That's a great idea.

This will speed things up radically because the thing about the large language models is they already have the statistical correlations built in. When you ask ChatGPT to generate 40 items that are conceptually like these four, that's what it does. It's not an opinion. So you can use ChatGP to purify the questionnaires. And you can do that on the left and on the right.

This will speed things up radically because the thing about the large language models is they already have the statistical correlations built in. When you ask ChatGPT to generate 40 items that are conceptually like these four, that's what it does. It's not an opinion. So you can use ChatGP to purify the questionnaires. And you can do that on the left and on the right.

This will speed things up radically because the thing about the large language models is they already have the statistical correlations built in. When you ask ChatGPT to generate 40 items that are conceptually like these four, that's what it does. It's not an opinion. So you can use ChatGP to purify the questionnaires. And you can do that on the left and on the right.

And it takes like 10 minutes instead of two years. I'm going to bring this back to Sonia.

And it takes like 10 minutes instead of two years. I'm going to bring this back to Sonia.

And it takes like 10 minutes instead of two years. I'm going to bring this back to Sonia.

We should be doing this with all the questionnaires. It's the same with narcissism. If you put... See, the other thing you could do with ChatGPT is you could say, here's 20 items significant of narcissism, okay? Which is the central item? And can you generate 20 items that are better markers of that central tendency?

We should be doing this with all the questionnaires. It's the same with narcissism. If you put... See, the other thing you could do with ChatGPT is you could say, here's 20 items significant of narcissism, okay? Which is the central item? And can you generate 20 items that are better markers of that central tendency?

We should be doing this with all the questionnaires. It's the same with narcissism. If you put... See, the other thing you could do with ChatGPT is you could say, here's 20 items significant of narcissism, okay? Which is the central item? And can you generate 20 items that are better markers of that central tendency?

And the thing is, it can do it because it's mapped the linguistic representations.

And the thing is, it can do it because it's mapped the linguistic representations.

And the thing is, it can do it because it's mapped the linguistic representations.