Stepos Venzny

I think this disparity in votes and comments is also hugely affected by how the UI has been changing over the years as well as the destruction of third party apps. The site is now designed in a way where active participation is less encouraged than ever before unless you’re running old reddit on a traditional computer with an ad blocker.

I will admit this is almost entirely gibberish to me but I don’t really have to understand. What’s important here is that you had any process at all by which you determined which answer was correct before writing an answer. The LLM cannot do any version that.

You find a way to answer a question and then provide the answer you arrive at, it never saw the prompt as a question or its own text as an answer in the first place.

An LLM is only ever guessing which word probably comes next in a sequence. When the sequence was the prompt it gave you, it determined that Homer was the most likely word to say. And then it ran again. When the sequence was your prompt plus the word Homer, it determined that Simpson was the next most likely word to say. And then it ran again. When the sequence was your prompt plus Homer plus Simpson, it determined that the next most likely word in the sequence was nothing at all. That triggered it to stop running again.

It did not assign any sort of meaning or significance to the words before it began answering, did not have complete idea in mind before it began answering. It had no intent to continue past the word Homer when writing the word Homer because it only works one word at a time. Chat GPT is a very well-made version of hitting the predictive text suggestions on your phone over and over. You have ideas. It guesses words.

deleted by creator

Let’s say hypothetically I had given you that question and that instruction on how to format your response. You would presumably have arrived at the same answer the AI did.

What steps would you have taken to arrive at that being your response?

But I don’t think the software can differentiate between the ideas of defined and undefined characters. It’s all just association between words and aesthetics, right? It can’t know that “Homer Simpson” is a more specific subject than “construction worker” because there’s no actual conceptualization happening about what these words mean.

I can’t imagine a way to make the tweak you’re asking for that isn’t just a database of every word or phrase that refers to a specific known individual that the users’ prompts get checked against and I can’t imagine that’d be worth the time it’d take to create.

Not in Incognito mode it isn’t.

Of course, one reason I might mind is if the machine uses what it learns from reading my work to produce work that could substitute for my own. But at the risk of hubris, I don’t think that’s likely in the foreseeable future. For me, a human creator’s very humanity feels like the ultimate trump card over the machines: Who cares about a computer’s opinion on anything?

This is really naïve. A huge number of people simply don’t care about creative works in those terms. We’re all encouraged to treat things as content to be consumed and discarded, not something to be actually thought about in terms of what it was expressing and why. The only value of a creator in that framework is that the creator fuels the machine and AI can fuel the machine. Not especially well at the moment but give it some time.

Because you have to have specific knowledge about how AI works to know this is a bad idea. If you don’t have specific knowledge about it, it just sounds futuristic because AI is like a Star Trek thing.

This current AI craze is largely as big a deal as it is because so few people, including the people using it, have any idea what it is. A cousin of mine works for a guy who asked an AI about a problem and it cited an article about how to fix whatever the problem was, I forgot. He asks my cousin to implement the solution proposed in that article. My cousin searches for it and discovers article doesn’t actually exist, so he says that. And after many rounds of back and forth, of the boss saying “this is the name of the article, this is who wrote it” and my cousin saying “that isn’t a real thing and that author did write about some related topics but there’s no actionable information there”, the boss becomes convinced that this is a John Henry situation where my cousin is trying to make himself look more capable than the AI that he feels threatened by and the argument ends with a shrug and an “Okay, then if it’s so important to you then we can do something else even though this totally would have worked.”

There really needs to be large-scale education on what language models are actually doing to prevent people from using them for the wrong purposes.