Look, AI will be perfect as soon as we have an algorithm to sort "truth" from "falsehood", like an oracle of some sort. They'll probably have that in GPT-5, right?

It's alright! There's a multibillion dollar advertising operation working to convince us that generative AI can do these sorts of things. Plus, it's always tough to go through someone else's work, much less mods from a decade ago that were written by ambitious amateurs. I couldn't read my own code after a couple of months if I wasn't such an absurd over-commenter.

If you want a chill intro to the real situation, I highly recommend this episode of On the Media that had Ed Zitron on. You could knock it out over a commute or two no problem: https://www.wnycstudios.org/podcasts/otm/articles/brooke-talks-ai-with-ed-zitron

A funnier and angrier thing that explains why this won't work can be found here: https://ludic.mataroa.blog/blog/i-will-fucking-piledrive-you-if-you-mention-ai-again/

No one:

Absolutely nobody:

Klarna: What if we financialized buying burritos using AI?

Our subjects here at awful systems can make us angry. They can spend lots of money to make everything worse. They can even make us dead if things go really off the rails, but one thing they can never do is make us take them seriously.

I think that what we've got here is a genuine victim of the hype.

Absolutely!

The thing about sov cits is that they use legalish words like a magical incantation. The words have no meaning to them, really. It's a tarted-up glossolalia which reifies their wishes to manifest some outcome in court.

If a lawyer surrenders their craft to a bullshit engine, they're doing the exact same thing: spouting law-shaped nonsense in the hope of getting the verdict they want, their only differentiator being that they showed up wearing a much nicer suit than the sov cit.

Surprisingly well-received by hackernews:

https://news.ycombinator.com/item?id=44053328

The main objection being along the lines of, "Milton Friedman's "I am not a racist" tshirt is raising many questions that are answered by his tshirt."

An image of a Github-themed restaurant that serves poop burgers.

I love how Wells has given us both a great series of stories AND a jokey terminator analog to diffuse the mAnLy trope of building and/or fighting terminators.

zbyte64 gave a great answer. I visualize it like this:

Writing software that does a thing correctly within well defined time and space constraints is nothing like climbing a smooth gradient to a cozy global maximum.

On a good day, it's like hopping on a pogo stick around a spiky, discontinuous, weirdly-connected n-dimensional manifold filled with landmines (for large values of n).

The landmines don't just explode. Sometimes they have unpredictable comedic effects, such as ruining your weekend two months from now.

Evolution is simply the wrong tool for the job.

Dad hi-five

IWW intensifies

A lawyer who depends on a sufficiently advanced AI is indistinguishable from a sovereign citizen.

If it makes you feel better, I've heard good folks like Emily Bender of Stochastic Parrots fame suggest confabulation is a better term. "Hallucination" implies that LLMs have qualia and are accidentally sprinkling falsehoods over a true story. Confabulation better illustrates that it's producing a bullshit milkshake from its training data that can only be correct accidentally.

It's soooo bad. Could it be a cry for help?

A reminder that AI slop tastes like boots.

https://old.reddit.com/user/Jackrobertsofficial/comments/1kklms1/hot_new_superhero_comic_swat_me_is_the_dirty_cop/?p=1

I didn't think I could be easily surprised by these folks any more, but jeezus. They're investing billions of dollars for this?

Self-licking ice cream cone As A Service

We have got to bring back the PE exam for software engineering.