Could ChatGPT be a viable way to answer people's questions?

Question

A few days ago, OpenAI, the people that brought you GPT-3 and DALL-E, launched a new product called ChatGPT. It can answer questions so accurately that it's legitimately scary. Even the most obscure topics known to man are answered swiftly by this bot. It's not 100% accurate, but it's shockingly accurate for an AI. To explain how accurate it is, someone posted on Twitter that they managed to take an SAT test with the bot and received a 1020, which is above the 50% percentile of all its test takers. People are calling this automated system as something that's better than Google.

My question is, could this be a way to accurately answer questions that are on any of the Stack Exchange sites? I entered a question from Stack Overflow relating to the NumPy Python module and it gave a pretty convincing response. The responses are professional, probably even better than similar sites such as Quora where the responses vary in quality.

Would a bot such as this one change the future of question and answer sites?

Answer 1

While the solutions often look correct to the untrained eye, they are frequently invalid in various subtle ways. As such they are actively harmful as they encourage the question poster to try things that simply don't work.

In one deleted answer the documented return value for COLUMNPROPERTY is either 1 or 2 but the chatGPT answer had code that was predicated on 3 being being returned, that code would never execute and therefore the answer didn't solve the question's issue. Worse than not working though was that it looked like it might be OK to someone unfamiliar with the possible return values of COLUMNPROPERTY.

When the question poster then comments that there is a problem with the answer, neither the answerer nor chatGPT is in a position to respond as neither understand the comment or what to fix in the answer to address it. I pointed out a number of such issues in my Stack Overflow answer

All known chatGPT answers on Stack Overflow have had to be deleted. When the post author didn't do it voluntarily, moderators have done it for them. Be warned that deliberately making work for moderators can result in suspension.

Having the same variable answer quality as Quora is not something to aspire to.

Note that using chatGPT to create answers on Stack Overflow is currently banned

Answer 2

I guess that a fundamental part of the question is what is the real value these answers bring. While in theory machine learning systems can string together plausible answers, a good chunk of what a good answer is involves experience and understanding. Many of these answers look good, but they are useless. One expects tasty cake, and finds frosting covered styrofoam.

Another thing is why are you here if your contributions are literally "type a question into a box, and blindly post the answer here". I've often talked of Stack Exchange as a commons, and essentially posting such answers is basically an act of vandalism. To borrow a term invented by Phillip K Dick, it’s kipple - useless detritus.

I'd argue that SATs, with a distinct, very focused domain and syllabus and a finite question bank, would also be different from Q&A where folks would have very arbitrary questions.

Basically, don't post solutions you didn't try or test yourself, or know for certain that they work, and something from a source like ChatGPT is suspect.

For factual questions – well, please confirm that well, the answer is factual, and not the fever dream of a hundred monkeys on a hundred typewriters.

Answer 3

It can answer questions so accurately that it's legitimately scary

It will be just as scary when it's wrong, but many won't realize until the consequences™ come.

Would a bot such as this one change the future of question and answer sites?

With the benefit of hindsight, to a degree, yes, of course, and to some degree, they'll continue to do so.

---

For context, this post's original revisions took a much more defensive stance, viewing the question from an angle of "how useful would it be for AI-generated answers be posted on Stack Exchange?" My previous answer focused heavily on need for quality in questions, and my defense for the continuing value of the Stack Exchange network was based on "garbage in, garbage out"- both for questions and source/training data:

• If someone doesn't know how to ask a good question, they're going to have a harder time getting a good answer from an LLM tool.

• For any new field with less information about it included in LLM training data and less SEO juice to get good info found in web search results, likelihood of a LLM + search engine giving useful results is probably going to be low.

A year has passed, and the aspect of this question that I find interesting has changed. Concerning the future of Stack Exchange, my new interest is in the net benefits this technology has provided / is providing- not on the whole world, but here.

---

could this be a way to accurately answer questions that are on any of the Stack Exchange sites?

Sure? (to an extent.) Nowadays, there are hybrids of search engines and LLMs, so all it needs to do is do a web search, and basically churn and regurgitate any existing answer content.

I've heard tons of people online and my friends say that they find technologies like ChatGPT to be a useful tool for getting answers to questions. Many of them who have newbie experience on Stack Exchange comment that they prefer it over Stack Exchange so they don't have to deal with grumpy experts.

That's what ChatGPT is good at: service with a smile, all the time.

People like that. Traffic has certainly declined (see Did Stack Exchange's traffic go down since ChatGPT?).

Okay, so how the heck do I see benefit in that?

Dealing with problematic question content is costly. It costs time, and it costs patience. It costs that to deal with questions that are missing key details, questions that are (for lack of better description) just a mess, questions that are duplicates, questions where potential for long-term value is covered in three feet of unnecessary context (and usually the hidden gem has already been asked before), etc. And each of those categories abound endlessly. Deal with it enough and it sucks the joy out of building a long-term knowledgebase.

ChatGPT and similar technologies do two things for that:

• It "tricks" people into doing research and basic troubleshooting.

  I say "trick", but really I mean that it masks research-related activities into a form that people find palatable. For some reason, many people either don't like to, or don't know how to google things (not that googling is a pre-requisite for asking a question on Stack Exchange (see How should we deal with Google questions? and Embrace the non-Googlers), but in my experience, it's usually (for question askers) a better search engine than the one built into Stack Exchange sites). But when the search engine is hidden behind a conversation with a machine, suddenly people find it very attractive (I guess the degree of synthesis and summarization (accurate/faithful or not) is also a factor).

  This isn't always a productive thing. We all know that an LLM can spit out falsehood as well as it can spit out truth. It can mislead people, and that can result in those people posting weird questions on Stack Exchange (Ex. I'm trying to do X and ChatGPT told me to do Y (which makes no sense)).

  But a lot of duplicates are very poor in quality, and less of those is a good thing. It means less curation workload, and less clutter mucking up searchability (bad signposts weigh out good ones). I've also started to see a (very small but hopefully increasing) trend in questions that show actually useful information prompted by conversation with an LLM. While this is slightly disappointing to me, since a lot of basic troubleshooting can be written statically and has no need for neural network involvement, it would seem that many of the people who love this technology are not the type to seek out the loving manual.

• It (to some degree) deals with the boring, obscure, one-off requirement dumps / questions that just require a lot of synthesis of basic concepts / questions with a bunch of unnecessary context.

  Or at least- that's what it seems to me like people who are trying to develop developer tooling around these technologies are trying to apply it to do.

  People are often too lazy to do or just don't know how to do is generalize their questions (de-localize / create things like Minimal Reproducible Examples) while not blowing up the question scope (staying focused), or to break them down into separate questions that are appropriately scoped for Stack Exchange. I don't get an awful lot of those questions in the tags I answer in, so I don't know how much this situation has changed, but I hope that among the classes of questions that are no longer being posted, there are fewer of these.

And in doing those boring things, it doesn't get grumpy. It has no social battery (well, it consumes a heck ton of energy to run, but that's a different can of worms).

I don't particularly care if ChatGPT wastes someone's time with a bunch of unhelpful info.

To the contrary, I welcome any side-effect that this technology has had on reduction in lower-quality question posts / increase in overall quality of question posts on Stack Exchange.

There are potential hidden costs- fewer people using the sites directly means fewer votes (and voting matters), and potentially missing some gems of questions and losing out on them to the machine. But a lot of people can't vote anyway, and I think if a question is useful enough, as long as this platform itself can retain its reputation as a trusted resource, the gems will find their way here.

Answer 4

"[ChatGPT] can answer questions so accurately that it's legitimately scary." It can also make things up, including citations to non-existent sources. Just say NO.

In response to something I wrote, it generated three paragraphs of politician-sounding pablum on a topic unrelated to my input (and the third paragraph was a repeat of the first).

Asked about the influence of Elvis Presley on the Sex Pistols, its response was 180° opposite from a public statement by Johnny Rotten.

It's been out less than a year, and I've already forgotten the other travesties I've seen. I consider it a minor improvement to an algorithm actually named "Travesty" from decades ago.

It has the potential to be a greater source of falsehoods than Facebook.

Answer 5

I would think there is definitely scope for AI to assist humans in providing good answers. Especially the latest version ChatGPT4 (with plugins like Wolfram Alpha for maths or ScholarAI or the web browser plugins to give sources) has very good performance quantified based on various benchmarks, e.g. ability to solve university entrance exams correctly (often top10% relative to avg. human) or maths ability. See my answer here giving some examples of how with the help of ChatGPT4 I was able to answer various SO questions, including ones that had never received an answer, despite being highly ranked and having been awarded 4 bounties before. Of course adequate advice on responsible AI use would need to be given, cf. the guidelines of many academic journals like PNAS on the use of generative AI. Especially for programming related questions I think allowing responsible AI usage could be beneficial. The guideline should then include that if ChatGPT or Github Copilot was used to develop code that reproducible code & a benchmark plus output should be given (confirming that the user verified the code works correctly, ChatGPT can't yet run code yet) & that use of ChatGPT should be disclosed & a reputable source should be given. That is very easy to check - much easier than telling if a given answer might have used ChatGPT if not disclosed. I would also not allow for plain copy & paste answers. So critical that appropriate guardrails to counter abuse would be in place. But I definitely see scope for humans & AI to work together to answer some of the questions on SO, especially when it concerns programming related questions (where reproducible code would make them easy to verify) and questions that never received an adequate reply. I would also suggest to add a checkbox "GPT assisted answer" so that user that would not like to see this content could still disable this if they would like. But why not make use of the free knowledge of user paying for ChatGPT4 & those that use it to assist in helping to answer questions on SO? If I look at the questions I ever asked on Cross Validated / Stats Stackexchange there is only 8 out of 20 questions that ever received an adequate answer, 3 of which I provided myself. If I paste the remaining unanswered questions in ChatGPT4 I would have checked 83% of the answers as being mostly or entirely correct (i.e. I would have marked them as correct accepted answers). With a bit of extra work these answers could surely be improved a bit further & it would be valuable to have them as correct, vetted answers.

Also, let's not forget: all those that having Bing as their search engine now by default get a GPT4 produced answer to their web search queries. So I would think that perhaps for most answers GPT now at least provided some inspiration for any answer... The important part is just teaching users how to use this technology responsibly & provide clear guidance on how to use it, as well as to point out its limitations. Imho, instead of fighting AI, humans and AI should work together to lift their abilities to a new level and allow them to solve problems they couldn't solve before. In the same way that we don't expect mathematicians or physicists to solve everything with pencil and paper and stop using computers.