Comments by "Winston Smith" (@winstonsmith935) on "JD Vance: These ideas will destroy western civilization" video.
-
9
-
@freewheelinflash Well I have tested piles of Comments over 30 days, and have throughly tested Googles/ YouTube AI, my conclusions.
That’s exactly the issue—AI can inherit biases from the data it’s trained on and from the people designing it. If an AI is trained primarily on sources that lean left, it may reflect those perspectives, and the same applies if it’s trained on right-leaning sources. The challenge is ensuring balance, but even defining “balance” can be controversial.
Many AI companies claim to aim for neutrality, but their definition of neutrality often aligns with the prevailing cultural and political views of those designing the models. Since tech companies are mostly based in Silicon Valley, where left-leaning politics dominate, critics argue that many AI systems reflect that bias. On the other hand, right-leaning groups have also explored creating AI models tailored to their viewpoints.
The key questions remain:
• Who decides what data AI should be trained on?
• How can bias be measured and minimized?
• Should AI present multiple perspectives on controversial topics?
Ive been in this game a long time, ever since i used my first computer in 1978
7
-
2
-
1