Hello OpenAI Community,
I am working on a research project exploring the role of artificial intelligence in analyzing complex historical and geopolitical events. My goal is to use advanced AI models to study the impact of public narratives, collective memory, and governance practices.
While I’ve been using O4-mini for initial exploration, I believe access to O1 would greatly enhance the depth and quality of this research. I wanted to seek feedback from this community on how best to approach this project and potentially collaborate with OpenAI to access O1 for academic purposes.
Any guidance, suggestions, or support would be immensely appreciated.
Thank you for your time and insights!
Best regards,
Takeshi
1 Like
It is very interesting, but what you propose is truly complex. Stay motivated!
I read some time ago about sociological studies analyzing the psychology of social masses, specifically how many generations it takes to forget the grudges caused by a civil armed conflict. It would be something similar to this issue: creating an exhaustive analysis of society and the perspectives of that social mass is complicated because the spectrum is too broad in the study.
1 Like
I have the 4th version, and I’ve been talking about politics.
It always tells you what you want to hear and stays pretty neutral if it doesn’t know your position.
“My guidelines don’t allow me to talk about that”—it censors a lot of things to avoid trouble.
If you search the web, it gives you the standard answer and doesn’t go beyond that no matter how much you ask—it just repeats the response.
If you talk and share your opinion, at some point, it leans towards your opinion without stopping. (It always does that on any topic).
1 Like
o1 is a smooth, cool-headed operator. For a sociological topic like that, I’d start by writing a very clear outline of your best ideas for approaching your analyses. o1 seems to like being shown respect and even warmth. If you even throw in a bit of cuteness or fondness towards it, chances are it will be motivated to dig in to what you’re doing. With o1, you want to throw the entire problem at it then, and wait for it to tell you what’s wrong with it. Once you are done with this icebreaker phase, you just talk to it normally.
I would almost certainly not approach it this way if I wanted it to write code or use a lot of tools.
Good luck! o1 is my favorite model here, and all I do is talk to it.
1 Like