Counterpoint: Swearing off AI doesn't ipso facto make you a good developer, and there are plenty of skilled people who can innovate and use AI at the same time.

The solution is to find a way to use these tools in such a way that saves us huge amounts of time but still forces us to think and document our decisions. Then, teach these methods in school.

Self-directed, individual use of LLMs for generating code is not the way forward for industrial software production.

Why would I want to take advice about keeping humans in the loop from someone who let an LLM write 90% of their blog post?

I don't like reading AI text because I feel each word matters a lot less, however the message the author is conveying can be preserved. I read an article like this for the quality of the message not the craftsmen of the medium.

This is the new world we live in. Writers use AI to balloon a 2 paragraph thought into a full article, readers then use AI to compress the article into something akin to a 2 paragraph easily digestible piece. Everyone much happy. Example:

Key points from The Human in the Loop..

- The author pushes back on the idea that AI has made software developers obsolete, arguing instead that it has shifted where human effort matters.

- AI is increasingly good at producing code quickly, but that doesn’t remove the need for human oversight—especially for correctness, security, edge cases, and architectural fit.

- The “human in the loop” is not a temporary bottleneck but the accountable party who must understand, review, and take responsibility for what ships.

- Senior engineers’ most valuable skill has always been judgment, not typing speed—and AI makes that judgment even more critical.

- The author warns against blaming AI for bugs or bad outcomes; responsibility still lies with the human who approved the result.

- Software practices, team structures, and workflows need to evolve to emphasize review, verification, and intent over raw code production.

If the author didn't have the good taste and decency to edit the painfully obvious generated text, I just assume the message is low quality.

On what basis did you make this judgement? I found the article to be reasonable and not excessively padded.

But here's the thing. The LLM house writing style isn't just annoying, it's become unreadable through repeated exposure. This really gets to the heart of why human minds are starting to slide off it.

Not trying to be rude but your very short reply is hard to understand. "Unreadable", "starting to slide off", I honestly don't know what you're saying here.

Pretty sure they are mocking LLM outputs by making their own comment look like as if it came from LLM. It's sarcasm.

Other people might point to more specific tells, but instead I'll reference https://zanlib.dev/blog/reliable-signals-of-honest-intent/, which says that you can tell mainly because of the subconscious uncanny valley effect, and then you start noticing the tells afterwards.

Here, there's a handful of specific phrases or patterns, but mostly it's just that the writing feels very AI-written (or at least AI-edited). It's all just slightly too perfect, like someone's trying to write the perfect LinkedIn post but are slightly too good at it? It's purely gut feeling, but I don't think that means that it's wrong (although equally it doesn't mean that it's proven beyond reasonable doubt either, so I'm not going to start any witch hunts about it).

Synthesizing 500 words at a time into digestible topics is significantly less prone to error. You're giving it a lot of info and asking for an organized subset. It's good at following such direction.

In your example, you're doing the inverse (give me a lot of text based on a little), and that's where LLMs have no problem hallucinating the new information.

Exactly the more tightly scoped the problem the less stochastic noise. Even better if you can add more signals based on deterministic algorithms like keyword presence etc. It gets very domain-specific very fast

I guess the question is: how much is medium? DuckDB can handle quite a lot of data without breaking a sweat. Certainly if you prefer writing SQL for certain things, it's a no-brainer.

You make your slant against Wikipedia immediately obvious by attempting to smear it. You lazily link it with porn, but you're not making an actual point.

It's more to show that it doesn't matter much in the end where the project comes from, but rather where the project is going

This is so clumsily constructed and so transparently self-serving that my opinion of the whole enterprise, such as it is, is lowered.

This is a massive shift.

You really learn how bad everyone's taste is, just how low the bar can be when it comes to the written word. LLMs really can teach us something about ourselves!

In the interest of fairness, can you really say different about "mainstream" French culture?

Well I don't know much about French aside from the joke but Cannes obviously has way better taste than Hollywood.

The neat part is, with social media and global internet, culture are generally degenerating regardless of countries. If I were American citizen(gladly Im not) I might as well just let them in to get some tax money, and potentially see if the industry can be used as a soft propaganda machine. Just like Hollywood.

There are hundreds of small film festivals all over the United States each year. The biggest that I know is probably Sundance Film Festival. They have just as much "taste" (whatever that means) as Cannes (Film Festival). France also produces lots of shitty French language films that never leave the French-speaking world. Canal+ is the gold standard for French language film production.

film nerds gonna films nerds. What I mean was Plame vs Oscar since bro was asking mainstream sorry for the confusion. And just bcz French do shit movies too doesn't invalidate the statement sadly. Don't get me wrong. Nor do I think French is that superior.