Getting Bad Vibes from
Simon Willison, who is often viewed as a “moderate” in the field of AI-assisted programming (he’s not, more on that in a moment) and thus is regularly cited by people who really want to convince you how concerned they are with the “age of AI” while also appearing “reasonable” (apologies for this run on sentence!) is now big mad that some book authors are misusing the term vibe coding.
From his blog post Two publishers and three authors fail to understand what “vibe coding” means:
Vibe coding does not mean “using AI tools to help write code”. It means “generating code with AI without caring about the code that is produced”. See Not all AI-assisted programming is vibe coding for my previous writing on this subject. This is a hill I am willing to die on. I fear it will be the death of me.
Well I’m sorry Simon, but you don’t get to die on this hill.
People love to throw around a dreaded phrase which I abhor — “the genie’s out of the bottle” — but in this very particular case I will deploy it. The genie’s out of the bottle, Simon. People are vibe coding for the simple reason they are reaching for generative AI tools and they trust them to be useful to them in some way they can’t be useful to themselves. That’s it. That’s all it has ever been or will ever be.
Again, from Simon’s article:
Vibe coding is when you forget that the code even exists, as a fun way to build throwaway projects. It’s not the same thing as using LLM tools as part of your process for responsibly building production code.
Here’s a hill I’m willing to die on: there is no process for responsibly building production code using LLM tools. And this is where we get into the vital discussion around what’s considered “moderate” in this space.
Simon is not a moderate. Simon is an extremist. He believes genAI-assisted coding is how programming gets done today. That itself is an extreme position. However, he happens to be less extreme than the really extreme extremists who outrageously believe coding as a profession will be replaced by “agentic AI” — therefore it provides the illusion of reasonableness.
The actual moderate position on this topic is that today’s genAI tools are dangerous and unethical and should be actively avoided.
That statement may come as a surprise to you, but if so it only serves to highlight just how insane the discourse is at this juncture. OK, so if Simon is on the kinda-but-not-totally-extreme pro-AI side, and I’m a moderate, what’s the extreme anti-AI side?
Committing violent acts to destroy physical infrastructure powering AI services. In other words, guerilla anarchism. An insurrection in the name of saving humanity and the environment.
Perhaps this too may shock you. Have you ever considered anarchism as a possible course of action? I have. In fact, if this AI bubble doesn’t burst soon due to natural causes because the hold of the technocratic oligarchs proves too strong, I fear this is an inevitable outcome.
Let me be very clear. I do not advocate for violence. I am a pacifist at heart, and I believe we will accomplish more through peaceful yet “loud and proud” advocacy along with class solidarity. Regardless, Generative AI is tied at the hip to fascism (do the research if you don’t believe me), and it pains me to see pointless arguments over what constitutes “vibe coding” overshadow the reality that all genAI usage is anti-craft and anti-humanist and in fact represents an extreme position.
We desperately need to reset the terms of the conversation, because the tech media by and large is too cowardly (or too dependent on their corporate overlords receiving major ad revenue by the very peddlers of genAI) to frame the narrative accurately. Again, for the folks in the back:
- Extreme pro-AI: using and promoting genAI tools.
- Moderate: choosing not to use genAI tools and warning others about it.
- Extreme anti-AI: attacking the physical infrastructure of genAI companies.
If you feel the need to scoff at this, asking me “where are these mythical so-called freedom fighters attacking AI companies?”, all I will say is we haven’t yet arrived at that point in the story.
I expect the first stage we will see is network-based attacks. GenAI services will increasingly be targets of hacker groups trying to take down the system. There will be PR campaigns both for and against these actions. I sincerely believe this is not a matter of if, but when—unless we see the bubble burst first, possibly later this year.
But to sum up, this is the kind of conversation we should be having around the topic of AI-assisted programming or “vibe coding”—not parsing the laughable semantics of exactly what that means.
Who cares?
It’s like an ostensible lung doctor who’s oh so very keen on finding the best way of managing smoking cigarettes so that the smoker receives the most benefit out of smoking cigarettes while minimizing the documented likely side effects of getting lung cancer and finally death.
Uh, maybe let’s not do that? At all? 😂
Enjoy Email Newsletters?
“Cycles Hyped No More”
Continue Browsing: May 2025
🚫 
🚫
👈