[The one about the day the dinosaurs died][2] was cool.

But I now know Claude can do these and more posts in this vein feel spammy and content-farmy.

Maybe link a blogpost or video of how you do/prompt these and then move on.

[1]:https://news.ycombinator.com/item?id=49835099

[2]:https://news.ycombinator.com/item?id=49841912

I mean, there are people with aphasia and hyperplasia - how do you prove actual 3D visualisation.

There are games that people play that need 4D visualisation to be able to do well and they perform at that.

Agreed, I don't think anyone has a good grasp on, say, 64K D, but we can do a bit better than just 3.

I have found that often LLMs are familiar with the contents of SF books.

But I feel poorer, almost deprived, by the fact that all the LLMs I've checked with have NOT been trained on the contents of Eon by Greg Bear.

Isaac Su seems to be irritated and mildly triggered that people don't think or act like LLMs aren't just dumb, mute, simple tools.

Instead of talking about THAT, he is doing a weird combination strawman/bikeshed-like post about the different imagined, nuanced reasons why he imagines people credit LLM input in attribution.

A more honest take would be:

"Some of you are ascribing more sentience and agency to LLMs than I feel comfortable with. Stop doing that. I'm obviously and fundamentally right that they are dumb tools that can never threaten my intellectual and emotional security or human superiority."

Whether they are right or wrong — one reason that many people are giving credit is that they feel that (at some at level at least) that Claude/DeepSeek/Gemini/etc. is acting enough like a participant or partner to deserve being given credit.

Stuff interesting enough to get upvoted (i.e. not slop), but I'm so irritated and triggered by pointless, human, comment threads moaning about "this is AI generated".

Wasn't too distracted or annoyed by the obvious AI voice.

Learned a bit about ice cream and American (Western?) enshitification at work...

...but these endless, superior, contentless, dogmatic, boring streams of comment threads about how AI it all is.

*Rolls eyes*

I almost actively _long_ for the days when commenters simply didn't read the article/post/treatise/repo and went straight to the comments, or complained about the paywall or the fact that it is on Twitter.

Please be interesting.

> but to say you’ve singularly thought through every problem in an armchair, in comparison with the distributed minds of the rest of us, is pretty arrogant.

Somehow incredibly ironic when applied to you comment itself...

I was on the fine tuning team of a multi-team hackathon to make a specialized chatbot once a few years ago and despite working technically well our output had very little impact on end to end output.

Technical curiosity?

I ask because this does not seem to be something to want to have.

These humanoid robots are expensive. Unskilled labour, housework seems to be the lowest ROI for them.

Factory work, dangerous tasks or skilled tasks I can understand that the robot can pay for itself, but you'd wear out the machine before paying off its labour on work like this.

I get the future angle for usage after population collapse means we need more bots to take care of elderly populations, but NOW?

Housebots don't seem practical.

Wow

See https://taalas.com/products/

Edit: updated link

1) >I simply run it through an LLM,

2) >paste the summary,

3) >and ask them "Is this an accurate summary?"

4) >and then I ask the for their original prompt.

Agreed that just step 1 or step 1 and 2 would be depressingly pointless, but step 3 and 4 make this the equivalent of sending someone a let-me-google-that-for-you kind of link, does it not?

Caught out like this I imagine many people will kind of get the fact that you'd rather have their direct inputs..

(Or just get mad at you, but that's fair I guess)

People who failed just did it wrong. /s

>

> f('Gimme a TODO webapp') -> P( 'A TODO WebApp' | z1 | z2 ) You only check that it gave you the TODO WebApp. Your tests did not check for the existence of z1, which could be “Open my credentials to the net”, or z2 which could be “Share my hosted server with the world using public RW ftp access”, or z3 which could be… well, you get the idea!

This is true when using compilers as well, right?

See Reflections on Trusting Trust" by Ken Thompson

Umm. Nope: see GCC flags.

In general, reproducible builds are possible, but takes significantly more effort.