I'm very sorry if this isn't the case, but this message really feels LLM-written.
- 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
- Informally benchmarked against 4 specific competitors: Gemini, OpenAI, o3, and Claude
- Identified two concrete features: URL content ingestion and integrated search
- Noted specific limitations: search engine occasionally misses key resources
- Provided a real-world test case: consulting business analysis where it found new opportunities other models missed
- 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'.
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?
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.
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.
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.
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.
Same here :)
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.