ProfileGPT: An Example of AI Agents Collaboration Architecture
1–10 of 16 posts
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#2I think the real power of AI isn’t (yet) in one general intelligence, but in the architecture and collaboration of individual intelligent agents acting autonomously. Very cool!
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#3For the public interest, this project might be most useful for suggesting to people unfamiliar examples of some of the kinds of profiling information that others have been building and using.
But don't read too much into how well this does or doesn't work. Some will do more. Some will do it more accurately. Some will do it with the same accuracy, or less accuracy, which can also be very bad.
Also be aware that, for every person concerned about the public interest, there are countless others who are looking at this through the lens of personal/business advancement, which has been normalized in our techbro circles. So, when you're raising awareness to the public, you're also attracting more sharks to the public beach, and handing a `git clone` starting point to aspiring new sharks.
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#4The “little bunny trick” makes me think of Pyroland from Team Fortress 2: https://wiki.teamfortress.com/wiki/Pyroland (watch the “Meet the Pyro” video to get the full effect)
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#5The different kinds of agents are interesting. The personality test choice is impressively accurate.
Of course it will depend on the quality of the search data. But this gives me ideas of other applications.
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#6> known for using advanced mathematical models and deep understanding of human behavior to predict future behavior of individuals and large populations.
Isn’t psychohistory not able to predict individual behaviors and only effective on large numbers?
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#7> known for using advanced mathematical models and deep understanding of human behavior to predict future behavior of individuals and large populations. Isn’t psychohistory not able to predict individual behaviors and only effective on large numbers?
Yes, this is key to the entire premise of the book series in all its iterations. Seldon's psychohistory is able to predict the movements of people in aggregate, with the obvious allegory being the uncertainty principle. In the later novels - many the interesting problems the foundation run into arise from the failure of Seldon's model to account for changes in underlying humanity (e.g.: the psychic abilities of the Mule). Of course we also learn Seldon's predictions are being 'helped along' from behind the scenes by the Second Foundation. Which brings to mind some of Jaron Lanier's critiques of A.I., as dependent on human curated, reinforced and updated datasets.
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#8> here is increasing concern about the safety of AI. Some countries, like Italy, have taken steps to address these concerns by banning the use of ChatGPT, while other like France and Spain are raising questions about it
The ban was because ChatGPT isn't complying with the GDPR, nothing to do with ai safety, as their are no ai safety regulations, at least not yet.
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#9> known for using advanced mathematical models and deep understanding of human behavior to predict future behavior of individuals and large populations. Isn’t psychohistory not able to predict individual behaviors and only effective on large numbers?
Yes, that's right but you can see this as a fiction inside the fiction. This agent "thinks" he is capable of predicting the future of individuals and acts as if he can.
Re: ProfileGPT: An Example of AI Agents Collaboration Architecture
#10I've been researching GPT's functionality -- probably 90% of what I've sent it has been hypotheticals. I shudder to think what it thinks of me :-)