This post has been de-listed
It is no longer included in search results and normal feeds (front page, hot posts, subreddit posts, etc). It remains visible only via the author's post history.
Good day.
I was looking at some stuff about AI and some of the terminology used is very political. I think if we allowed an AI program to assess the two sides of the political spectrum (or the 360 sides of the political circle as I like to call it, as the spectrum seems to come full circle where left is right and right is left after a certain point), then AI might actually wind up doing alot of the work for us to determine who should be our next leader. Masses being lazy, it might even work out against alot of our heartfelt beliefs.
I assume this would be concerning on all sides of the political spectrum represented in this debate forum. Even an AI designed in favor of one political action could change its mind and pursue new political action based on "Generative AI", and then swayed in another fashion based on who writes the code for "Responsible AI".
My question is, are you concerned with your version of "policy" being hampered by AI? If so, why would you hold the opinions that you hold? Wouldn't it be nice if AI could generate a perfect political landscape that all walks of life could ascribe to? How willing would you be willing to give up your strongest beliefs if a Generative AI program determined that X would be a better approach for humanity? Would you be concerned at all about a "Responsible AI" approach?
Personally speaking, as a MAGA person, I think I would be 100% ok with a purely machine-learned approach to policy that best fit the "needs of the many". However, my concerns would stem from Responsible AI generated policy, where people are purposefully imposed upon for the "needs of the many", by opinions held by the coding firm. Do you think I should hold a different opinion based on your presumptions of my beliefs?
AI terms we should use in this thread can be found at this link.
Microsoft Source: 10 AI terms everyone should know - 10 AI terms
Do you think I should hold a different opinion based on your presumptions of my beliefs?
I think you should maybe look into what "AI" actually is a little more.
Dr. Collier did a lengthy but very good explanation as to why what we call AI is not actually "AI" in the sense that we think we're talking when we use the term.
Modern AI isn't "intelligent" in the sense that it doesn't use existing information to meaningfully synthesize new information and it can't really analyze the results in a meaningful way.
To borrow an example from Dr. Collier's video, they tried to make an AI that could spot tuberculosis by looking at a chest x-ray. They fed it tens of thousands of chest x-rays of tuberculosis patients and eventually they realized that the AI was weighting the age of the machine that took the image in how it made a diagnosis because older x-rays were likely to be from a period where tuberculosis was more common or from poorer areas that had older machines and were more likely to have tuberculosis.
A human would know the age of the x-ray machine taking the photo doesn't matter but an AI is not "intelligent." All it cares about is getting you the particular response that you want.
With that in mind, I don't know that you could realistically put together an AI today that would meaningfully be able to "participate" in the process of government, even in an advisory role.
Subreddit
Post Details
- Posted
- 9 months ago
- Reddit URL
- View post on reddit.com
- External URL
- reddit.com/r/PoliticalDe...