The main problem I see with this right now is the fact that AI isn't actually coming up with any of these answers itself. Just like any other question, it is compiling all of the answers from anyone that it has access to their content and just giving you a summary or average of what those answers would be. It's all still really subjective and in my opinion, there is zero thought involved in it. Just like when you ask it what the price of BTC will be in 10 years. It just finds articles about the price of BTC in 2035 and then takes the average of those numbers.
What is your take on the phenomenon of the wisdom of the crowd, where the collective judgment of a group of individuals, especially when diverse and independent, is more accurate than the judgment of any single expert? It would seem that AI presents an averaged out judgement.
Not at all. Wisdom of the crowd assumes that the crowd has some knowledge that has value. Groupthink and herding behaviour, in other words, people who are homogenous - are not wiser together.
hence the part about diverse and independent.
Is AI diverse and independent?
Without knowing the source of the data that it used to average out a response, I couldn't say one way or the other. However, we have no reason to believe that AI replies are solely derived from single sources of data. And it's not much different from scholarly publications that often cite the same sources, yet we give expert status to people with this derivative work.
Remember when the AI images of the typical Nazi soldier came out a year or so ago, and half of them were black and ethnic? It isn't a derivative of scholarly articles, it is the derivative opinion of whatever is socially acceptable now - the crowds. It isn't always this clear though. The way it nudges opinion further is dangerous, isn't it?
I think that was a result of the programmers requiring AI to generate images with diversity. I think it's more reflective of the creators of the AI.
Still, I do agree that we should be mindful of accepting all that AI spits out. So long as it is a black box, we can never be certain about what biases it harbors.
No in fact it's the opposite as you know. Enter the echo chamber...
I think it's still too early to tell. Definitely not something I would want to bet my life savings or my mental health on. The problem with the current Internet is I don't think the points of information are diverse or independent enough to say one way or the other.
Certainly. And we should not forget that AI doesn't actually understand what it synthesizes. It is capable of seeing patterns that might escape people. But at the same time it will also reflect biases of the data used.
Very True!
For sure. But there is also zero thought in the people using it to solve whatever problems they are looking to solve. It is a spiral of thoughtlessness.
We are demons of convenience, even when it leads to our own downfalls.