Post

It’s also not akin to the antivax movement because not using AI doesn’t leave you and your community vulnerable to disease, like what are the risks of rejecting LLMs?

bsky.app/profile/rycp... The Purpose of Superintelligence or whatever is Not Having To Die And Experience Nothing Forever I'm under the impression that most people believe in An Afterlife, which because of my conditioning I interpret as saying Death Isn't Real

If I wanted to cut hair or do nails in the state of California I’d need a license. But I could be writing load-bearing code for a public utility and no one would ask me for one. This is actually at the root of a lot of our current problems.

Finally, "useful" is a very loaded term that doesn't mean much in the way that the AI boosters use it. I fundamentally do not give a shit about your subjective experience of whether you think the chatbot, which is designed to convince you that it is helpful, is helpful to you.

Thank you

For me, any estimate for how long I think it'll take me to do a project, I double or triple that estimate before I present it to anyone. And then I still get it wrong often and about half of the time it's non-coding stuff that I can't plan for that derails my estimate and half is weird bugs.

Do climatologists have accreditation? Both groups typically have degrees.

I mean, I'm a research software engineer among other things, guess how much people care when I point out massive problems with the current "AI" fad :)

The lack of duty of care, general disdain of history, and understanding the motivations (the “why”) of things is increasingly disturbing to me as years pass.

In my time at university, I had one (1) professor who was adamant that the customer is not always right, a software developer should have opinions and standards & should push back when asked to build stupid solutions. The fact that he was the odd one out for this was certainly a sign, looking back.

It really shouldn't be legal for comp sci / software dev to be called engineering.

This is a very important thing I wish more people understood. A very large part of the reason why SE/coders specifically have found so much use for LLMs is because their field of engineering is a uniquely lawless uncivilized dog-eat-dog wasteland where everyone keeps having to reinvent wheels.

I’ve been kinda distancing myself from this discourse since I don’t really have domain expertise, however so much of the arguments about what LLMs show about language & cognizance are just complete failures of 101 level epistemology & formal logic.

Software engineer here; I'll cosign that. You don't even need a specific degree for it; I know devs who when to business school and became "techies" purely bc it was easier to prove their own value as an engineer than as a consultant.

Then there’s the “all computer security professionals” claim, which he seems to be flat-out lying about.

I am a software engineer, and most software engineers I've worked with are bricks in the wall. Fittingly, every other role in software considers engineers the bricklayers of the team: fairly incompetent, utterly replaceable, and wholly without authority. Nobody listens to us, so neither should you

Or translating an entire codebase from one language to another

What would you prefer your proctologist's motto to be, "do no harm," or "move fast and break things?"

That's the worst part. LLMs and useful ML getting both mixed under the umbrella term AI. If someone says 'BuT aI iS hElPiNg To CuRe CaNcEr' to defend the current hype cycle they flat out lie, and I think this is on purpose because a layman can't fact check this easily.

spreadsheet people (laudatory) are exposed to its possible utility a lot more than humanities people, the later seeing most of its total egregiousness. And so when Pro-AI says it's here to stay, Anti-AI side hears that all the awful parts of AI is here. question is, good outweigh the bad?(hellno)