Unless I am wildly misreading this, this is actually worse that both GUIs and LLMs combined.

LLMs offer a level of flexibility and non-determinism that allow them to adapt to different situations.

GUIs offer precision and predictability - they are the same every time. Which means people can learn them and navigate them quickly. If you've ever seen a bank teller or rental car agent navigate a GUI or TUI they tab through and type so quickly because they have expert familliarity.

But this - with a non-determinstic user interface generated by AI, every time a user engages with a UI its different. So they a more rigid UI but also a non-deterministic set of options every time. Which means instead of memorising what is in every drop down and tabbing through quickly, they need to re-learn the interface every time.

I don't think you have to use this if it's not working in your case. I think the idea is to try to anticipate the next few turns of the conversation, so you can pick the tree you want to go down in a fast way. If the prediction is accurate, I could see that being effective.

It’s intended for conversations that are probably different every time too. It’s like a more expressive form of what Claude Code already does with the “AskUserQuestion” interface.

> GUIs offer precision and predictability - they are the same every time.

Except after an update everything is in a different place.

Yep - I'm looking at you MS office ribbon. Just as I learnt where things are some update decides to move stuff around.

The people responsible for stuff like this should be put in stocks in public squares and pelted with tomatoes ;-)

> If you are correct, that implies to me that LLMs are not intelligent and just are exceptionally well tuned to echo back their training data.

Yes.

This is exactly how LLMs work. For a given input, an LLM will output a non-deterministic response that approximates its training data.

LLMs aren’t intelligent. And it isn’t that they don’t learn, they literally cannot learn from their experience in real time.

There is some intellegence. It can figure stuff out and solve problems. It isnt copy paste. But I agree with your point. They are not intellegent enough to learn during inference. Which is the main point here.

Wolves (and all dogs) could be vegetarians as they aren't obligate omnivores - and in certain conditions where pray is sparse they do eat berries to surviven. Cats on the other hand are obligate carnivores and can't produce taurine amino acids, so they have to eat meat to survive.

We can chemically synthesize taurine just fine.

Are you a wolf (or a dog)?

I think the implications is that cats could eat veggies laced with synthetic taurine...?

I thought the implication is that people should feed themselves to cats?

Consent removes a bunch of ethical issues.

What's eating you, Earthman: https://m.youtube.com/watch?v=5HLy27bK-wU

Current state AI doesn’t have hands. How can it possibly be better at installing electrics than anyone?

Your post reads like AI precisely because while the grammar is fine, it lacks context - like someone prompted “reply that AI is better than average”.

An electrician with total knowledge/understanding, but only the average dexterity of a non-professional would still be very useful.

Finally, I can give up this life of a programmer and live my dream of being a moisture farmer.

What’s worse being accused of an AI post or being defended because your post is so bad that AI wouldn’t have written it?

I don't expect someone to do deep focused work from 9am to 5pm.

But at the same time, I don't expect them to spend their 9-to-5 working for another company at the same time.

As a founder, who respects the 9-to-5 and supports WFH, if I'm paying for 8 hours of work, I want 8 hours of output. Not 4 hours of output, and then you working 4 hours for another job.

If multi-jobbing becomes a thing, then WFH becomes untenable because at least in the office you can be monitored.

To be fair, you're either paying for hours or for output, because I assure you you are not paying staff accurately for their output. You can of course sack someone who outputs notoriously little, but if you get output exceeding your average "8 hours of output", you shouldn't care if someone made it in 1 hour or 16, or at least you wouldn't be able to tell.

I'm using "output" as quoted in context, it's such a nebulous measure unless you're specifically buying a product.

Do you pay your programmers hourly or on salary?

How do you measure whether some output corresponds to 8 hours of work, and not 4 or 16 hours?

He doesn't known what he is talking about. Bunch of wannabe founders waxing BS. If you want 8 hours of guaranteed output use a bot

But why is it that news online defaults to US politics?

Yes, politics impacts everything, which is all the more reason to exclude it, otherwise how do we draw the line?

Are we buying protests in Indonesia against disappearances of protestors under the rug but not posting about it? https://www.abc.net.au/news/2025-10-20/prabowo-subianto-one-...?

It does seem like a rather modest proposal.

> Have you considered that … conservatives are self-selecting out or can't cut it at all?

When the ingroup is underrepresented the question is “how can we get more ingroup in”.

But when the outgroup is underrepresented it’s met with “maybe outgroup just isn’t as good”.

Eg. women in tech, men in nursing; conservatives vs liberals in academia.

It’s worth thinking about why you’re wrong, because it explains why those comparisons aren’t valid. Modern conservatism is defined by rejecting ideas like objective truth or pluralism which are the core of academia. There’s no way to have conservatives more represented in science when conservatives refuse to allow people who practice science to be part of their movement: you used to be able to find Republicans who wanted to do something about climate change, for example, but anyone who wants to apply scientific principles there now has been purged from the party. Vaccines aren’t quite there yet but it’s trending in that direction and the percentage of doctors who are Republican have been declining since the pandemic.

Contrast that with women in tech or men in nursing and it becomes obvious why the comparison isn’t valid: women want to be good technologists, not to reject the validity of technology or say we should all go back to the Amish lifestyle (that desire is common to senior developers of both genders). Male nurses want to be good nurses, not claiming that their gender means they can commit medical malpractice.