It’s gracious of you to say that you’d be sorry, and I did run my comment through 4o (perhaps ironically) which caught a slew of typos and weird grammar issues and offered some improvements. But the robotic sound and anything else you don’t like are my own responsibility. Do you, perhaps, have any thoughts on the substance of the comment?
That's discomforting. My practice of sprinkling em-dashes like salt on a salad dates from my early days on various video game communities' forums. They comfortably mimic interrupted speech in writing. I hope I won't have to soon defend myself against accusations of AI usage just because I belong to the minority that read The Punctuation Guide[0] or a related resource.
It's really the em dash along with superfluous language. I suspect you are fine. Models like 4o have a very specific pattern when folks don't specify their writing style.
- Very 'forced' expressions (back-pocket tests, 'The analysis is razor-sharp')
- The fact you're glazing AI so much means you probably uses it, it's like how it was with crypto bros during all the web3 stuff
- Lack of any substance, like, what does that post say? It regurgitates praises over the AI, but the only tangible feature you mention is the fact it can receive an URL as it's input
Hmmmm it is hard to really place the issue. I am very much in the bullish on AI camp but I don't like writing for the sake of writing and some of the models (4o in this case) have very obvious tells and write in such a way that it takes away from what substance may exist.
One thing that concerns me is when you can't tell whether the comment was authored or just edited by AI. I'm uncomfortable with the idea that HN threads and reddit comments gradually tend towards the grey generic writing style of LLMs, but I don't really mind (save for the prospect of people not learning things they might otherwise!) when comments are edited (i.e. minor changes) for the sake of cleanliness or fixing issues.
I just re-read the post twice and I couldn't find any of the points you mentioned (again, other than using URLs in the input):
- Informal Benchmarks: I'm sorry, what? He mentions 'It’s picking up on nuances—and even uncovering entirely new angles—that other models have overlooked' and 'identified an entirely new sphere of possibility that I hadn’t seen nor had any of the other top models'. Not only it is complete horseshit by itself, but it does not benchmark in any way or form against the mentioned competitors. It's the exact stuff I'd expect out of a LLM.
- Real-World Test Case: As mentioned above, complete horseshit.
- 2 Concrete Features: Yes, I mentioned URLs in the input. I didn't consider 'Integrated Search' (which I'm assuming is searching the web for up-to-date data) because AFAIK it's already more or less a staple in LLM stuff, and his only remarks about is is that it is 'solid but misses sometimes'.
Its because of the em dashes (- is a normal dash, — is an em dash). Very few real people use those outside of writing books or longform articles.
There's also some strange wordings like "back-pocket tests."
It's 100% LLM generated.
What is much scarier is that those "quick reply" blurbs on Android/Gmail (and iOS?) will be able to be trained on your entire e-mail and WhatsApp history. That model will have your writing mannerisms and even be a stochastic mimic of your reasoning. So, you won't be able to even realize a model answered you, not a real person. And the initial message the model is responding to might be written by the other person's personal model.
The future of digital interactions might have some sort of cryptographic signing guaranteeing you're talking to a human being, perhaps even with blocked copy-pasting (or well, that part of the text shows up as unverified) and cheat detection.
Going even a layer deeper / more meta: what does it ultimately matter? We humans yearn for connection, but for some reason that connection only feels genuine with another human. Whereas, what is the difference between a human typing a message to you, a human inhabiting a robot body, a model typing a message to you, and a model inhabiting a robot body, if they can all give you unique interactions?
Everyone who uses a compose key has it available (via ---) — I do. You mean the em-dash though, not the en-dash, and Davidzheng is using hyphens for approximation, not en-dashes.
Not really, as pointed out by others in the thread. Anecdotal of course, but I use em dashes all the time— even in emails and texts (not just long-form writing).
I often write things I want to post in bullets and then have it formulated better than I could by an LLM. But its just applying a style. The content comes from me.
My wife is dyslexic so she passes most things she writes through ChatGPT. Also not everyone is a native speaker.
TBH I've recently felt like that for ~70% of 'top-level replies' in HN, which has slowly pushed me to other mediums (mastodon and discord).
Could just be that the AI 'boom' brought a less programming-focused crowd into the site and those people lack the vocabulary that is constantly used here, who knows.
I'd go out on a limb and say I think probably LLMs made the general population aware of how the "general voice" feels/looks/reads like.
So rather than a lot of people adopting to write like how a LLM writes, the LLM writes as an average of how people been writing on the internet for a long time. So now when you start to recognize how "LLM prose" reads (which I'd say is "Internet General Prose"), you start to recognize how many people are writing in that style already.
I've been in the internet since the early 2000s, I can assure you it does not write like how 'someone on the internet' would write. And when I say that, I mean that for both sides of the internet: it doesn't sound like how 'old school' internet folks would write, but it also doesn't sound like how teens talk either. Neither of these groups write in 'very plain' English regurgitating useless information.
Recent trends/metas in video formats like tiktok and shorts encourage that kind of 'prose', but I haven't seen it being translated into text format in any platform, unless it's written by LLMs.
My point wasn't that it writes like any specific groups, but a general mix-match made up of everyone voice, but a boring average of it, rather than something specific and/or exciting.
Then of course it depends on what models you're talking about, I haven't tried Grok3 myself (which I think you're talking about, since you say "it"), so I can't say how the text looks/feels like. Some models are more "generic" than others, and have very different default prose-style.
