Comment by gwern | Hacker News Reader

Comment by gwern | Hacker News Reader

Parent Full thread

Likely AI-written. OP sanitized some of the most obvious tells, or perhaps the blogging software added those spaces to the em-dashes, but when a dev who is as gung ho on GenAI as https://chrlschn.dev/blog/2024/10/im-a-gen-ai-maximalist-and... is, writes an ending like that, and the whole thing is about as intellectually nutritious as a bag of Doritos, we all know what they did. A waste of time.

---

Out of curiosity, I asked ChatGPT-o3 what LLM it thought wrote the ending, and I agreed with its rationale for guessing that 4o wrote it:

"The prose is archetypal ChatGPT rhetoric—second-person apostrophe (“dear traveler”), triadic repetitions, and a moralizing close. / Vocabulary inflation (“complexity merchant”, “moat of self-importance”), syntactic parallelism, and the cadence of stacked, comma-separated clauses are hallmarks of GPT-4’s temperature-controlled completions. / No concrete anecdotes, data, or links—just abstract exhortation—exactly what an LLM defaults to"

But more interesting was the rationale for it not being Claude or Gemini-pro: "those models tend to hedge (“may”, “might”) and inject more safety language. This text is bolder and more florid—typical of GPT-4 with style-boosting prompts."

I'm not sure I would've identified that as stylistic tics of Claude/Gemini. I'll have to look closer for those in the future.

We ask people to avoid making accusations like this in the comments, because the hurt and damage caused by a false accusation outweighs the benefit to the community of making the accusation publicly.

Instead, please flag the post and email us so we can take a look and see if we think the content and the comment thread are HN-worthy, as that's ultimately all we can assess. The originality and quality of the writing is always an important consideration, regardless of how it was created.

(We also discourage pasting LLM-generated output into comments. In this case it doesn't really prove anything either way and takes the comment further into off-topic territory. More broadly, we want to keep HN for discussion between humans.)

I am the author, my blog is an Astro template and you can see the GH repo here for this post: https://github.com/CharlieDigital/chrlschn/commits/main/src/...

And of course, the full commit history of all of my writing: https://github.com/CharlieDigital/chrlschn Even a simple look would have led you to the repository with full commit history of my fixes for grammar, spelling, etc.

I've also kept a blog long before GPT existed: https://charliedigital.com [0] so I would point you to that and my long history of writing (e.g. https://chrlschn.medium.com/) that this is hand rolled content. Rather I'd say AI has copied my style of writing.

Please feel free to peruse and use your bulletproof analysis to let me know which of these spanning back to 2005 are AI written. I hope that your takeaway from this is that your AI spidey senses are completely broken and realize the irony of your intellectual laziness here relying on an LLM instead of just following the links in my profile or clicking the link to my GH on my blog. If you have any sense of righteousness, please retract your flag and let the mods know of your error.

[0] Exaxmples:

https://charliedigital.com/2020/11/13/principles-of-speed/

https://charliedigital.com/2020/10/29/how-leaders-fail-their...

https://charliedigital.com/2020/09/20/in-praise-of-simplicit...

https://charliedigital.com/2016/04/11/recipe-crunch-time-exe...

Reply on news.ycombinator.com