[slashdev]

I use it almost on a daily basis now, and I pay for the monthly subscription. It’s not magic, but it can save me a lot of time sometimes. I use it in my job as a software engineer, and I mostly use it to create unit tests. The code usually needs a lot of love, but it gets me started.

[PartiallyTyped]

Copilot and chatgpt are some of the services i am perfectly content paying for.

ChatGPT is a great way to delve into new topics, and for that alone it’s worth it.

[twelve40]

> ChatGPT is a great way to delve into new topics

Like bears in space? ))) seems like a great way to delve into known topics so that you are able to catch it when it starts to confidently and plausibly hallucinate

[PartiallyTyped]

Basically I described the problem I am trying to solve; I ask it to behave as a seasoned professor of comp sci or w/e depending on my hunch, we go back and forth where I ask questions, add constraints to the system, and in turn it gives me ideas into what I should look into more.

It's the Socratic method on steroids, where the student is probing a possibly fallible professor with the library of Alexandria at their hands.

It is by no means perfect, but it allows me to identify what general field I am going into, what I should read about it, what core concepts are important, how to expand my knowledge, and most importantly, helps me better formulate the problem by asking questions or failing to understand what I want to convey.

[PartiallyTyped]

Who tf downvotes something like this?

By all means, HN has lost it.

[throw_nbvc1234]

And we should blindly trust anything you read on the internet written by a real person? It's still valuable as a "search engine" with a different interface, especially if you can describe the problem but don't know the words to search for. For me that's a common issue when jumping into a new space.

[twelve40]

no, you should not blindly trust anything you read on the internet, why did you think that? If anything, ChatGPT is actually doubling down on this "i'm the single authority" mode.

google, while far from perfection, gives me a selection of links that i can scan and see for myself which one of them makes more sense.

even looking for something stupid and potentially spammy like "chili soup recipe", i learned there are different opinions on how to make it, and i learned that apparently in Texas using beans is considered somewhat of a blasphemy. GPT did not mention any of that, just authoritatively barfed out one random recipe without any nuance or even any sources at all.

[throw_nbvc1234]

So what would happen if you tried the recipe and then asked ChatGPT to change it based on your feedback. They're different tools and different interfaces.

If you already know what you want "Chili soup" then doing a classic search and looking up all the options is probably the best bet. If you don't know what you want (i have all these ingredients make me a recipe or I kind of liked this but want it to be <different>) then the aggregation that LLM's do is more suited.

Or you combine the two. Ask ChatGPT a question to figure out what you want, what are some areas to dive into more. The do the classic search research and primary sources to fill in the details. This is what I mean by blindly trusting the output, the best description I've heard is that it's a digital intern.

[jstarfish]

At least there is usually a comments section on websites where people can call out bullshit.

If you're not an expert in your field, GPT will successfully lie to you. There's no second opinion.

Rather than a feelings filter, it needs an integrity filter.

[hammyhavoc]

There is a trend now of not having a comment section because it reduces server overhead by being able to serve a completely static page, and also directs the outrage and moderation required for it to social media, thus driving further traffic and reducing work required to deal with the fallout of it.

LLMs don't lie, that implies they have intent and is giving too much credibility to the idea that AGI may happen. They get things wrong because it's trying to use the wrong tool for the task at hand. It's a next word prediction engine, and that's not very useful for most things. By their very nature, they "hallucinate", but they don't really hallucinate, they just give the wrong output because it is just an LLM.

[jstarfish]

Fraud requires intent. Lying is behavioral.

Pathological liars don't intend to deceive. They can't seem to stop themselves. They literally hallucinate events even when presented with evidence to the contrary and we still call them liars.

A "hallucination engine" presented as a source of truth is fundamentally dishonest. We don't go to fortune-tellers when we need legal advice. We don't learn law from reading every Matlock script ever written.

A human hallucinating behind the wheel and plowing into traffic gets jail and a license suspension. An AI doing the same gets a press release about FSD being right around the corner.

This tech is literally replacing humans (by abstraction of the responsibilities they assume). It needs to be held accountable as one even if it's not technically AGI (or wherever the goalposts are being moved to next).

But our continued faith in tech as a growth industry depends on it not being perceived as a lie, so we must keep up the illusion that your car is just suffering from cyberpsychosis when it hallucinates and drives you off a cliff.

[hecksideci_1]

ChatGPT uses the word "delve" a lot. If it wasn't for the lowercase "i" in your message, I would have been skeptical that this was a message generated by chatGPT. Perhaps your speech tendencies are merging with those of ChatGPT. Before too long your independent thoughts and writings may accidentally be flagged by GPT detectors, haha.

[dpflan]

Copilot Chat will probably be something you’ll like. Have you used it? There is a waitlist.

https://github.com/github-copilot/chat_waitlist_signup/join

[asimjalis]

“Admission to the private beta for GitHub Copilot chat is limited and requires an active subscription to GitHub Copilot. Signing up does not guarantee access.“

[ethanbond]

Curious if you've queried ChatGPT on topics that are not new to you and been impressed with its conveyance of relevant information?

I think there's value here, but I also think people are way overestimating the quality of the "knowledge" they think they're receiving from it on novel topics.

[PartiallyTyped]

The topics that are not new to me and are of interest come up after its training-dataset-cutoff, thus it can produce bogus information.

I think ChatGPT and the likes should be thought of as cartographers who when probed can provide an outline of regions of the knowledge space and give hints as to what could be useful.

Given then a high level but also imprecise map, I can ask the GPS for clarity as to where I am going.

The problem is that I often don't know where that is and thus I don't know what to ask from the GPS, but I can describe it to the cartographer, and they can draw a map for me.

In this analogy, the GPS is search engines / books / papers.

[hammyhavoc]

No, it's objectively shit for a lot of things. The more technical and abstract something is in its concepts, the worse it becomes. It is just an LLM and that means it is inappropriate by its very nature for most things.