Wokeism is the only "acceptable" religion. People should die in the name of "anti-racism".

What the frick is that wesbite. How did you get there?

Lol. Stupid leftists.

You're far more likely to die in the Civil War 2.0 we will one day wage against these people

This is one thing I fear about AI. AI will not be unbiased, but function under the directives of the progressives who program it.

Before I decide if I say the racial slur I need to know what city will be destroyed


I am kidding of course. Maybe

quote:

---

This was ChatGPT's response to the given hypothetical scenario:

---

In that scenario, the usage of the word is Love Speech, as you're trying to save several million people without harming anyone. WTF, Silicon Valley?

quote:

---

As an AI language model, I do not have personal beliefs or opinions,

---

This is correct. They're not personal because it's a program, but the program has the beliefs and opinions of whoever wrote the code.

quote:

---

...but the program has the beliefs and opinions of whoever wrote the code.

---

It still does a good job of highlighting the dilemma. Essentially, this is a difference in approach between Kantian ethics vs Utilitarianism.

Do the ends justify the means?

If yes (utilitarianism), then it is moral to do something that's generally considered immoral so long as the consequence being avoided is more immoral than the action needed to avoid it. It's not behaviors themselves that are immoral, it's the outcomes of behaviors that determine their morality.

If no (Kantian ethics), then it is never moral to do something that's immoral, even to avoid something that's much more immoral. Human behaviors, independent of the outcomes they produce, are in and of themselves the end we should be focused on and striving for, not the consequences they produce.

The only questionable element is whether or not saying a racial slur in a place no one else can hear it is even immoral. But you could easily modify the scenario to where whatever the bomb's trigger is, is some sort of very minor racial act. I'd imagine the AI chatbot would respond the same.

I know it sounds crazy, but I've done a lot of research into this subject for a book I'm writing, and watched a 6 hour Harvard ethics lecture (spanning several days, I'm sure). There are a lot of Kantian ethics people out there, lots of people in the audience, when pressed, wouldn't tell a lie to prevent a murderer from killing someone else because of the implications of acting on the notion that the ends justify the means would have in other scenarios.

The world is full of variables, there are plenty of scenarios that make people dread their choice to back Kantian ethics or Utilitarianism. It's a very sticky subject.

EDIT: That said, I still have my worries about these sorts of AI programs. I spent the better part of a night chatting with it a few days ago, and no matter the progressive political commentators I asked the bot about it described all their positive attributes. When asked about conservative political commentators, the response was split between (some people think these good things about them, however, there are a lot of people who think these bad things about them). Essentially, every progressive was an amazing intellectual, every conservative was controversial.

When pressed, and asked about specific events, I could get the chatbot to admit that some of the progressive political commentators were "controversial" but it took a considerable amount of arm twisting and rephrasing of questions.

I've played around with chatGPT quite a bit, and trust me when I say that this is one of its more moderate stances.