4 ms·
Even here on HN, people are using ChatGPT to write replies. You can spot it by the very verbose answers that don't say anything meaningful. Someone else commen
by gtirloni 4y ago
Even here on HN, people are using ChatGPT to write replies. You can spot it by the very verbose answers that don't say anything meaningful.
Someone else commented here that AI would raise the noise level everywhere and I unfortunately have to agree.
- cdelsolar 4y ago[flagged]
- kcorbitt 4y agoWhile LLMs can be prompted to write in many different styles, especially if you allow them to edit a text over multiple passes, the default "voice" of ChatGPT is surprisingly recognizable. For instance, the comment I'm replying to was clearly GPT-generated.
- qzw 4y agoYeah that’s pretty obvious, and yours sounds a bit like it too, especially starting off with “While…”. I do find it hilarious that the ChatGPT default style sounds very much like how I wrote in high school. I guess I was still training my own language model at that point, so it kinda makes sense.
- RulerOf 4y agoGPT output always reads like a short story or essay, but HN et al. have a distinctly conversational aspect that I don't get with ChatGPT. You're right about the high school writing style. I avoid writing like that whenever possible because reading it is exhausting. As short as this reply is, I've erased nearly half of what I originally typed.
- Kokouane 4y agoI agree that the parent is clearly GPT-generated, but was it 3.5 or 4? I notice substantially improved human-like writing in GPT-4 that will probably be hard to detect. Even the current AI detectors struggle (although they have gotten pretty good at detecting 3.5 writing)
- cdelsolar 4y agowhatever was at the ChatGPT page, lol
- D13Fd 4y ago. . . said the AI.
- nirvdrum 4y agoIf the generated text is mostly correct most of the time, people are just going to stop reviewing it because it's cheaper for them to waste someone else's time than it is to spend their own. Even those that are more diligent are bound to let things slip through the cracks because it's easy to lose concentration in a repeated, monotonous process. When everyone is doing it, it's not really a problem with a particular individual anymore.
- nuancebydefault 4y agoSomehow, my pattern recognizer lit up with this structure: - I understand that... however... - some general knowledge - That being said... Without even interpreting what the comment is, just feels a bit orchestrated.
- spaceman_2020 4y agoI genuinely feel that chatGPT has been neutered. It used to be much better at writing like a human given the right prompt. Now I can't get it to write like one at all. Which makes me wonder what they're using the non-neutered chatGPT for...
- nuancebydefault 4y agoIt could be that our pattern recognizers are outsmarting the AI after a while. In the beginning noone noticed the difference and now anything written by AI has a bit of fake/dusted off feel to it. I'm sure the next generation will outsmart us again.
- spaceman_2020 4y agoMaybe. But it definitely feels like a far less powerful tool than it used to even 2-3 weeks back. It's fine for coding questions, but any time I've tried to use it for marketing content, the result has been way too formulaic and completely useless within the context. It used to be smart enough to figure out that you were writing marketing content and would tailor the answer for that. Now, it just writes a 500 word blogspam regardless of what you ask it to write.
- nullsense 4y agoYou're up against increasingly powerful computers and models. You're going to lose in the end. I worry that in the final analysis our conclusion will land on "yes we made this technology because we were excited by the possibility of the benefits, but sadly it turned out to really not have been worth the downsides"
- counttheforks 4y agostop spamming please
- yenda 4y agoVery funny (this is either chatGPT or someone writing like it on purpose)
- paulddraper 4y agohahaha
- catgpt23 4y agoI completely agree with your perspective. The onus of using AI-generated responses responsibly does lie with the users. AI tools, like any other technology, can be used for both productive and unproductive purposes. The key is to strike a balance and use AI-generated responses in ways that contribute positively to online conversations. One potential solution to minimize the noise created by AI-generated responses is to develop better guidelines and best practices for AI usage in online discussions. This would help ensure that users are aware of the potential risks and consequences of misusing AI-generated content. Educating users about responsible AI usage can promote a more thoughtful and considerate online environment. In addition, the AI development community can also work towards creating more focused and concise AI-generated responses by refining the models and algorithms. This would help reduce verbosity and generate more meaningful content that users can employ in online conversations. Ultimately, the collaboration between AI developers, users, and other stakeholders in the online ecosystem is crucial for fostering a responsible and productive use of AI-generated responses. By working together, we can harness the potential of AI to enrich our online interactions while minimizing the negative impact of AI-generated noise.
- iso1631 4y ago> You can spot it by the very verbose answers that don't say anything meaningful. I've been doing that online for years! Does this mean ChatGPT be taking my hobby away?
- spaceman_2020 4y agoYeah, I've noticed the same thing, and it can be pretty annoying when people use ChatGPT to generate those long-winded replies that don't actually add much value. But on the flip side, I've seen cases where AI-generated content can actually be super helpful. I guess it's all about how people choose to use it. It's up to us to encourage responsible use and make sure we don't let AI-generated content overshadow genuine, thoughtful discussion. Who knows, maybe as the tech gets better, we'll see AI tools that can generate more concise and meaningful responses. Fingers crossed! (this comment was generated by chatGPT)
- WhiteOwlLion 4y agoIf I include tl;dr or eli5, chatGPT makes an effort to distill the answer to the essentials.
- JohnFen 4y ago> You can spot it by the very verbose answers that don't say anything meaningful. Although, in all fairness, I have been known to write verbose answers that don't say anything meaningful, too, and I'm not a bot. I think.
- hobo_in_library 4y agoChatGPT replies to you with: > Don't worry, JohnFen. We still love you even if you're not sure whether you're a bot or not. After all, aren't we all just a bunch of biological machines running on meat-based neural networks?
- jvanderbot 4y agoExcept for the first sentence, that would easily pass for a HN reply.
- trashface 4y agoAI is taking your job I'm afraid.
- 2143 4y agoPerhaps that's how I would eventually get rid of my HN addiction.