The thing is, no one has the slightest idea how to stop hallucinations.

The models are fundamentally "hallucinatory" at core - they generate what is _probable to follow the string thus far in its training corpus_, modulo RLHF and friends.

Notice that nothing there has any rigorous relationship to truth.

Sure, the companies could start pumping money into pure research on what models other than transformers might yield something that can reason rigorously, but at that point you're talking about finding a way to throw out LLMs entirely in favor of a less-pathologically-broken model, like Gary Marcus keeps complaining people should be doing.

When management at $DAYJOB brought the hammer down and said, "Everyone has to use genAI all the time, OR ELSE," I expected to be blown away by the tool I was avoiding due to ethical concerns, aesthetic objections, humanism, and long-term thinking.

I was born away, but not in a good way.

The CLI is _bad_. I've seen it randomly fail to render anything at all on the terminal multiple times. It has a vim-mode, but it's painfully buggy, and I can literally outrun it - if I try to type too quickly after hitting Esc for normal mode, it just doesn't return to normal mode. It's I was keeping track of the bugs in the Claude TUI, but gave up because it was taking _too much of my time_ to do so.

If nothing else, I'd say Claude shows convincingly that success is not the default for vibecoding.

Yes, it technically does the job, and no, I don't think I've ever used a worse TUI.

I've considered keyboard-driven WMs before, but have thus far gotten by using regular WMs with a few custom keybinds for window sizing, movement, and monitor placement.

As long as you have an automated algorithm for the basic layout, it works pretty well.

I still have to contendb with apps that don't offer the shortcuts I want, though, and that's when Shortcat is very handy.

As one who's driving Claude daily due to corporate mandates, I can see why people fall in love with genAI coding, but my revulsion has only grown as I've learned to do it, so I won't be spending my free time with LLMs.

The name is very reasonable, but effectively ungoogleable.

https://shortcat.app/

Similar to Vimium, but for the whole OS. Apparently Homerow is similar, judging from comments I'm seeing here.

I really wish I knew an equivalent for Linux. I might even leave Gnome behind if a different DE has a good model for this.

I suppose poor implementations of boring old deterministic spell-checking are a thing, too.

Maybe that's some part of the disconnects in this thread.

Knowing how something works is not the same as having the tools to change it.

Discovering memories are incorrect does not massively change who we are. As someone with a very defective memory, I discover on an hourly basis that I'm won't about something I thought was true, but there's still continuity and consistency to my personality and general approach to life.

...in fact, as someone who was raised an evangelical Christian and believed wholeheartedly without a shadow of doubt, then lost my faith entirely in my late thirties, I sort of did have my "introductory paragraph" changed, yet my wife, children, and friends would all say I'm still me, and that my core personality and nature remains largely the same.

> Are you aware of the Recovered Memory Therapy Scandals of the 80s/90s ? Boy did that ruin a lot of lives. You can rewrite a human by changing their 'introductory paragraph'. It's just not as accessible.

The recovered memory scandals are not even close to evidence that you can rewrite a human.

The people who thought they had learned new facts about themselves did not suddenly lose their context as humans in 20th century America.

They did not suddenly lose their sense of humor, or develop a previously-unseen penchant for murdering small children.

They experienced a revision of belief, and a pretty major one that really distressed them, but it did not change everything about them.

LLMs _do_ manifest wildly differently based on the first paragraph.

If your goal is to check your writing for plausibility and rough grammatical correctness, that's certainly an open problem for deterministic, conventionally-written software tools.

My goal with spell checking is to make sure my occasional mechanical typos while using a desktop computer get caught before someone else has the chance to be annoyed by them.

I don't have an issue with using the wrong word entirely when writing at a computer, so that's not a use case I think about. It does happen when I use a smartphone, due to autocorrect and predictive typing, but that's not a case this Claude skill applies to.

So, for my use case, the ~6 orders of magnitude more energy used to send documents over the network to be hyperchurned on an array of GPUs guzzling electricity is pure waste.

It also makes the whole process orders of magnitude _slower_.

I find that massive waste and slowdown infuriating, even while conceding that it can perhaps deliver a little more value then the deterministic spell-checking algorithms I rely on.

I wanted to hurl my laptop out to the window.

Not unlike Buckethead, they seemed to prefer avoiding the direct spotlight of fame, but they still wanted to create these things and put their thoughts, feeling, and original aesthetic out there in the world.

...all of which is to say "except for the AI part" is a pretty big exception.

We have no such evidence that LLMs do.

That's a pretty significant difference between the next-token predictor and the squishy lump of meat.

I haven't looked at how long he's been predicting job destruction, so I don't know if that explanation fits the facts.

FTFY.

...though even the billionaires may regret it if the peasants wind up starving by the billions. Species don't do well with tiny populations.

And before anyone says "UBI," give me a coherent explanation of:

* who is going to fund UBI to the tune of fifty trillion dollars * why we're so confident they'll do that * why there are currently so many people starving and homeless in SF if if any tech billionaire feels the need to spend their money providing for other humans

My experience so far is that it's harder and slower for me to understand the genAI code than to write it myself.

Skipping thorough comprehension seems to be the popular choice in my workplace, but it's not one I can justify.

I'm not sure, but it seems to me that if scale or small architectural tweaks were going to solve comprehension, they would have done so by now.

What do you base this on?

As someone who can both see the amazing things genAI can do, and who sees how utterly flawed most genAI output is, it's not obvious to me.

I'm working with Claude every day, Opus 4.7, and reviewing a steady stream of PRs from coworkers who are all-in, not just using due to corporate mandates like me, and I find an unending stream of stupidity and incomprehension from these bots that just astonishes me.

Claude recently output this to me:

"I've made those changes in three files:

- File 1

- File 2"

That is a vintage hallucination that could've come right out of GPT 2.0.

I don't see how this could be achieved.

Any widely-recognized benchmark is going to be gamed by the genAI companies.

They have a strong financial incentive to do so, and their products' nature shows that they are not influenced by ethical or societal-good incentives.

Makes it clear you used a bullshit machine, without implying it's an author.

...assuming you think using them at all is a good move - I won't deny they have some utility (though I'd argue much lower than many seem to think), but I do presently believe they're a disaster for humanity.

The ruination of the Internet with slop, the massive propagation of propaganda, and the insanely easy-to-wield tools for abuse are in no way worth the ability to accrue tech debt at 10x velocity (though to be clear, accruing tech debt can absolutely be a useful strategy, if one I personally dislike).