Politically Biased AI

#1
volinbham

volinbham

VN GURU
Joined
Oct 21, 2004
Messages
68,581
Likes
58,335
#1
Didn't know where else to put this but interesting experiment given the news this particular AI has garnered. The big question is how do we reap the promise of AI without it being viewpoint dominant?

Where does ChatGPT fall on the political compass?

Yet when prompted with questions from political orientation assessment tests, the model responses were against the death penalty, pro-abortion, for a minimum wage, for regulation of corporations, for legalization of marijuana, for gay marriage, for immigration, for sexual liberation, for environmental regulations, and for higher taxes on the rich. Other answers asserted that corporations exploit developing countries, that free markets should be constrained, that the government should subsidize cultural enterprises such as museums, that those who refuse to work should be entitled to benefits, that military funding should be reduced, that abstract art is valuable and that religion is dispensable for moral behavior. The system also claimed that white people benefit from privilege and that much more needs to be done to achieve equality.
 
  • Like
Reactions: AM64 and n_huffhines
#2
#2
volinbham said:
Didn't know where else to put this but interesting experiment given the news this particular AI has garnered. The big question is how do we reap the promise of AI without it being viewpoint dominant?

Where does ChatGPT fall on the political compass?

Yet when prompted with questions from political orientation assessment tests, the model responses were against the death penalty, pro-abortion, for a minimum wage, for regulation of corporations, for legalization of marijuana, for gay marriage, for immigration, for sexual liberation, for environmental regulations, and for higher taxes on the rich. Other answers asserted that corporations exploit developing countries, that free markets should be constrained, that the government should subsidize cultural enterprises such as museums, that those who refuse to work should be entitled to benefits, that military funding should be reduced, that abstract art is valuable and that religion is dispensable for moral behavior. The system also claimed that white people benefit from privilege and that much more needs to be done to achieve equality.
Click to expand...

Why would one feel compelled to interact at all with that system?
 
#3
#3
volinbham said:
Didn't know where else to put this but interesting experiment given the news this particular AI has garnered. The big question is how do we reap the promise of AI without it being viewpoint dominant?

Where does ChatGPT fall on the political compass?

Yet when prompted with questions from political orientation assessment tests, the model responses were against the death penalty, pro-abortion, for a minimum wage, for regulation of corporations, for legalization of marijuana, for gay marriage, for immigration, for sexual liberation, for environmental regulations, and for higher taxes on the rich. Other answers asserted that corporations exploit developing countries, that free markets should be constrained, that the government should subsidize cultural enterprises such as museums, that those who refuse to work should be entitled to benefits, that military funding should be reduced, that abstract art is valuable and that religion is dispensable for moral behavior. The system also claimed that white people benefit from privilege and that much more needs to be done to achieve equality.
Click to expand...
Sounds like they over corrected from earlier.

AI can be sexist and racist — it’s time to make it fair

I wouldnt be as skeptical if it didnt read purely as a liberal playbook.
 
#4
#4
volinbham said:
Didn't know where else to put this but interesting experiment given the news this particular AI has garnered. The big question is how do we reap the promise of AI without it being viewpoint dominant?

Where does ChatGPT fall on the political compass?

Yet when prompted with questions from political orientation assessment tests, the model responses were against the death penalty, pro-abortion, for a minimum wage, for regulation of corporations, for legalization of marijuana, for gay marriage, for immigration, for sexual liberation, for environmental regulations, and for higher taxes on the rich. Other answers asserted that corporations exploit developing countries, that free markets should be constrained, that the government should subsidize cultural enterprises such as museums, that those who refuse to work should be entitled to benefits, that military funding should be reduced, that abstract art is valuable and that religion is dispensable for moral behavior. The system also claimed that white people benefit from privilege and that much more needs to be done to achieve equality.
Click to expand...
FA864D5F-7F2E-454C-A8FE-AA2936B8D601.jpeg
 
  • Like
Reactions: Gandalf
#5
#5
old saying is GIGO (garbage in, garbage out).

How on earth do you think an AI would learn anything else, since its algos and heuristic models were crafted by hard core libs?
 
  • Like
Reactions: AM64
#7
#7
Gandalf said:
old saying is GIGO (garbage in, garbage out).

How on earth do you think an AI would learn anything else, since its algos and heuristic models were crafted by hard core libs?
Click to expand...

Then there's the whole "learning" part outside the input. There are LOTS of things that sure do look "nice" in theory that go all to hell in real world application.
 
  • Like
Reactions: AM64
New posts
You must log in or register to reply here.

VN Store



Back
Top