Hacker Newsnew | past | comments | ask | show | jobs | submit | ffsm8's commentslogin

> There is no moral argument anymore.

There never was.

There were just a few people profiting from ads trying to gaslight you into believing there was.


> new JS framework of the month

I got news for you: that hasn't been the case in 10 yrs now. It's no longer 2010 in case you didn't notice.

Seriously, react, emberjs angular, vuejs etc - they're all essentially the same age at this point.


Isn't tanstack the new hotness now? Or is it Vite? Or is it htmx?

Most Jobs I've worked at would've forbidden the use of personal subscriptions like that, as you'd be effectively uploading their intellectual property to foreign actors.

This can end with more then a termination - as in being literally liable/on the hook for serious contract violations.

So ymmv, you may want to take care with such an approach


I didn't mean to do it sneakily, I meant to be straight up with the employer and offer to pay for the subscription if they won't.

Huh, we have a process which has several exit criteria - which are pretty expensive to calculate with multiple rest calls to get through each...

I've always called them exit hatches, entirely unrelated to llms...

Now I wonder if I need to reword the docs... But realistically speaking, llms are the only readers of them nowadays, so I guess it doesn't matter.


People usually make the determination by reading at least part of the text and then find multiple smoking guns / llm-isms

The comment you responded to did not have those.

Fwiw, the article we're commenting on was likely not LLM written. The sentence structure is too convoluted, no LLM would've generated it like that - unless very carefully prompted ... But at that point it's no longer pure AI slop (imo).


Isn't that precisely the reason why we introduced the term hallucination? Because llms have historically always made up bullshit of they cannot answer directly... If they now nailed this to maybe the model not respond instead of responding incorrectly, then a lot of previously unusable usecases would become feasible.

So I feel like that's exactly the right metric and the way to track it wrt hallucinations.


I had a buddy in high school that was notorious for doing the same thing. (He's now a senior director at a Big 4 consultancy. :) )

Do you mind expanding a little more?

They had a buddy who used to lie a lot when they were younger… now they get paid for it

The point is that it's not a useful metric on its own. For example, redirecting from /dev/null also achieves a zero hallucination rate.

We want the hallucination rate to decrease while the overall answer rate of queries remains sufficiently high. For more specifics, look into ROC and AUC.


Uh, aren't you confirming his opinion with that? After all, Anna doesn't have the money to fight this in court

No. Anthropic fought and paid $1.5 billion in settlement and agreed to delete all the copyrighted material.

I'm confused here, how is this not even more of a confirmation?

Essentially: have funny amounts of money and the law ceases to matter. Or don't, and be squashed by the right holders


$1.5 billion is more than $19.5 million though.

Also if I were to guess the damages because of sci hub is higher than Anthropic training the models. I don't think I know anyone who didn't bought a book because the summary is available or they can ask about it to AI.

Delete? Wasn’t that material already used to train models?

All AI companies should be forced to re-train their models without the offending materials, and this should also extend to all LLMs distilled from models exposed to copyrighted works. Also cover code under licences such as GPL as well. Not to mention patents and designs. This whole LLM business is a giant IP laundromat.

One of the best things about it IMO -- or we'd be spending the next hundred years waiting for copyright reform

"Deleting" data they already ingested is meaningless.

well i guess its copyright not distill-statistical-model-from-it-rights.

It's almost certainly ai written though. All the regular tells are there... Though he likely edited some out, like that "just"

Also if it was handwritten, it'd have been a third in length, the rest was LLM fluff


Correct, that was my point

I see, i actually like these tells. It let's us easily distinguish garbage from someones thoughts.

And you can also see how brainrotten someone's gotten when they start accidentally sneaking in these tells into their normal communication.

As a matter of fact, after a full workday in which I'm essentially forced to read LLM garbage for 9h a day... I sadly notice myself adding the same fluff pointlessness to how I express myself. like I caught a viral contagion that's actively siphoning my humanity away.

And expectedly, when coming back to those opinions with a less infected mindset, I frequently have to reevaluate these thoughts later on


Didn't you mean Claude take? It's ai written after all...

The 512 GB ram studio can't even be purchased anymore. It's been delisted

https://www.apple.com/shop/buy-mac/mac-studio

Same with the Mac mini. entirely removed from all store references


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: