Discussion about this post

User's avatar
Simon's avatar

I have spent months doing what you are doing on Chat GPT. I also have read up on it's creation and programming and "guardrails".

The similar, biased type answers that clearly support gender ideology are NOT from the AI. They are part of a surrounding protective bubble that the company (Open AI) has put on top of Chat GPT. So when you ask any questions that are about certain topics about groups of people it trots out these "canned" answers and actually does not let CHAT GPT actually answer the question using it's AI and learning.

They are doing this because in previous attempts to make AI language models the AI would veer off into non woke, non approved narratives and say things that data supported but that the elites considered racists, bigoted, transphobic, etc.

So, you see, who is controlling the AI and the guardrails around the AI will determine the answer. There are non woke AI being developed as well by other organizations that align more with free speech, but the big boys such as Microsoft and Google and Facebook will keep their AI from telling the truth in order to stay within the approved narratives.

Expand full comment
Average Dad's avatar

How romantic. I don't feel bad for AI not able to make sense out of this non-sense. It is what I would expect. I feel bad for the thousands of people affected by this terrible idea pathogen named Transgender.

Expand full comment
34 more comments...

No posts