I’m sorry but I’m frustrated by the blatant misuse of AI by my students and colleagues alike. It’s so obvious when they don’t understand what they’ve written.
That’s unfair to microwaves.
That’s unfair to microwave ovens because they have established uses, even in some fine dining establishments. So-called AI has none of that just yet.
AI has many very well established uses
Not really, it has a couple of niche uses mainly because people externalised the cost of coming up with a good analytical solution to their data processing problem (e.g. medical imaging analysis) which would be vastly more efficient and give insight into the underlying mechanisms, but that would cost grant money rather than VC capital and further externalised energy and environmental costs which are finally born by us, the taxpayers. Ultimately the technology as a whole is delivering very little value and like all hype bubbles mainly serves as a way of further enriching billionaires. But text generator go brrrrr
So many you didn’t list one.
Also OP didn’t talk about AI broadly, just vibe coding.
Microwaving is cooking. Vibe coding is to microwaving what staring at the food and pretending you have heat-ray vision is to microwaving.
don’t understand what they’ve written
Well first of all, they didn’t write it.
I would say it’s more accurate to say that vibe coding is to coding what microwaving a ready meal is to being a restaurant chef.
is vibe coding just using an llm to assist or whatever?
It’s not just using an LLM to assist. It’s more generating the whole source with an LLM, running it once to check if it seems to work (if it “vibes” good) and then publishing it without even trying to read through and understand the code.
Edit: just to clarify, the odds are that the generated code performs awfully, doesn’t handle even the simplest edge cases and has security problems.
