Hacker News Reader

Comment by InsideOutSanta | Hacker News Reader

I'm not sure why you would instruct an LLM to reason in this manner, though. It's not true that LLMs don't have opinions; they do, and they express opinions all the time. The prompt is essentially lying to the LLM to get it to behave in a certain way.

Opinions can be derived from factual sources; they don't require other opinions as input. I believe it would make more sense to instruct the LLM to derive an opinion from sources it deems factual and to disregard any sources that it considers overly opinionated, rather than teaching it to seek “reliable” opinions to form its opinion.

View on HN

>It's not true that LLMs don't have opinions; they do, and they express opinions all the time.

Not at all, there's not even a "being" there to have those opinions. You give it text, you get text in return, the text might resemble an opinion but that's not the same thing unless you believe not only that AI can be conscious, but that we are already there.

As a rebuttal, I offer a hacker koan: In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6.

“What are you doing?”, asked Minsky.

“I am training a randomly wired neural net to play Tic-Tac-Toe” Sussman replied.

“Why is the net wired randomly?”, asked Minsky.

“I do not want it to have any preconceptions of how to play”, Sussman said.

Minsky then shut his eyes.

“Why do you close your eyes?”, Sussman asked his teacher.

“So that the room will be empty.”

At that moment, Sussman was enlightened.

You're just using a different definition of "opinion", one that is too reductive to be useful in this case. If an LLM outputs a text stream that expresses an opinion, then it has an opinion, regardless of whether it is conscious.

“Opinion” implies cognition, sentience, intentionality. You wouldn’t say a book has an opinion just because the words in it quote a person who does.

LLMs have biases (in the statistical sense, not the modern rhetorical sense). They don’t have opinions or goals or aspirations.

Biases can lead to opinions, goals, and aspirations. For example, if you only read about the bad things Israelis or Palestinians have done, you might form an opinion that one of those groups is bad. Your answers to questions about the subject would reflect that opinion. Of course, less, biased information means you’d be less intelligent and give incorrect answers at times. The bias would likely lower your general intelligence - affecting your answers to seemingly unrelated but distantly connected questions. I’d expect that the same is true of LLMs.

Biases result in opinions.

Reply on news.ycombinator.com