Damn. Please never stop writing
It got "100% AI" with "high" confidence, highlighting that it was likely "paraphrased or rewritten" [1].
I know Pangram gets a lot of flak, but so far I've found it to be pretty reliable, especially since they claim to have a very low false-positive rate.
I then compared it with the original version of the article published on the author's GitHub [2], but got the same disappointing result [3].
And now I'm stuck. I don't know whether I should believe Pangram or the author of the article...
[1] https://www.pangram.com/history/38ffe92b-0077-4bce-816a-b89b...
[2] https://github.com/Suranjandas7/public_writing/commit/6b08ad...
[3] https://www.pangram.com/history/e9fea5bf-a2cd-4ba2-b6b0-75e0...
I suppose it helps to focus on the core of someone's strength, and the core of the content they are sharing.
You could be a great engineer and a great blogger, but have no skill in making graphics. If you made the artistic choice to add AI-generated art to an excellent engineering blog post, I would forgive it -- assuming it wasn't detracting in some important way.
Overall, the human skill of editing - particularly excluding that which is not additive - does remain key though.
I will say in this case, I don't know the pop song well enough to know if the AI-generated image is super relevant, or just random.
EDIT: I'm guessing it's "Kant"[0] in the image, listening to modern music. Ah... at first I thought it was like, George Washington.
The real problem is that "serious content creators" believe (as far as I can tell anyway) that they must have a cover image to make the article appealing or set the mood or whatever. They might even be right as far as the actually reader is concerned, but either way this is the real disease, and it's decades older than LLMs.
Anyway, it is their piece, their website, so they choose how to structure it. I just don't like the plague of obligatory featured images everywhere in general. I can understand a featured image that adds something relevant, or even simply has some artistic value in itself. But both stock images and AI-generated ones add neither.
Also, the author claims:
> No AI was used to generate the text for this article. The cover image and the voiceover in the audio version of the article is AI-generated. The cover image is made using Google Gemini’s latest image model while the voiceover uses OpenAI’s tts models.
You can slap an arbitrary disclaimer on anything; it doesn’t mean it’s genuine.