HackerTrans
TopNewTrendsCommentsPastAskShowJobs

nojs

3,359 karmajoined قبل 6 سنوات
[email protected]

Submissions

Ads are coming to AI, but not to Claude [video]

youtube.com
1 points·by nojs·قبل 5 أشهر·1 comments

Ask HN: Aggregated Weekly World News?

1 points·by nojs·قبل 5 أشهر·1 comments

Coffee Sniffers

en.wikipedia.org
3 points·by nojs·قبل 8 أشهر·0 comments

AI Darwin Awards – 2025 Nominees

aidarwinawards.org
3 points·by nojs·قبل 9 أشهر·0 comments

Rue Merdiere

languagehat.com
2 points·by nojs·قبل 9 أشهر·0 comments

Mycorrhizal Network

en.wikipedia.org
13 points·by nojs·قبل 10 أشهر·0 comments

comments

nojs
·أول أمس·discuss
> people said this would be too expensive

I imagine this is why the filter is so bad. Doing it with an intelligent model that better understands intent would be too expensive, currently.
nojs
·قبل 6 أيام·discuss
Would you really write “Private video titles aren't just metadata”?
nojs
·قبل 12 يومًا·discuss
Just want to say I really enjoyed your writing style, it’s just the right amount of funny/witty without distracting from the (very interesting!) ideas.
nojs
·قبل 27 يومًا·discuss
> all be it

fyi you probably mean “albeit”.
nojs
·الشهر الماضي·discuss
[dead]
nojs
·الشهر الماضي·discuss
“tell my obnoxious boss to fuck off about the tps reports” isn’t a great career move for them though
nojs
·الشهر الماضي·discuss
that was a fantastic story, thanks
nojs
·قبل شهرين·discuss
It’s in the image, designed to survive those kinds of operations
nojs
·قبل شهرين·discuss
One solution I haven’t seen recommended much is to have a Claude instruction/skill that explicitly audits the diff of every upgrade, and force this manual audit as part of your upgrade workflow. This seems like it would work pretty reliably.
nojs
·قبل شهرين·discuss
> $300/day token quota

Are companies using per-token billing? Why - is there some reason they can’t buy the $200/mo Claude plan for every employee?
nojs
·قبل شهرين·discuss
What about access to GPUs and memory? This is becoming a pretty major bottleneck.
nojs
·قبل شهرين·discuss
> Every week, somewhere between 1.2 and 3 million ChatGPT users, roughly the population of a small country, show signals of psychosis, mania, suicidal planning, or unhealthy emotional dependence on the model.

> Why is mental-health crisis not a gating category, the kind where the conversation stops, full stop, and the user is routed to a human?

Well, obviously “routing to a human” is not feasible at that scale. And cold exiting the conversation is probably worse for the user than answering carefully.
nojs
·قبل شهرين·discuss
Yeah, the solution given is actually wrong as stated!
nojs
·قبل شهرين·discuss
This conflict is popping up everywhere. There is a push by a lot of companies to allow agentic use of their services (and new companies explicitly offering "X for agents"), ignoring the fact that "agent" means the same thing as "bot" which we've spent the last couple of decades actively filtering out. Will be interesting to see how it plays out.
nojs
·قبل شهرين·discuss
Another vote for this - we’ve been using it for years without issues.
nojs
·قبل شهرين·discuss
[flagged]
nojs
·قبل 3 أشهر·discuss
My working theory is that all models are approximately the same, and the variance in quality mostly depends on how long they think for.

So the trick is to always set to max, and then begin every task with “this is an extremely complex task, do not complete it without extensive deep thinking and research” or whatever.

You’re basically fighting a battle to make the model think more, against the defaults getting more and more nerfed to save costs.
nojs
·قبل 3 أشهر·discuss
> Ads do not appear in accounts where someone tells us—or we predict—they are under 18.

Time to make a deal with the kids - i’ll verify you for instagram if you verify me for ChatGPT
nojs
·قبل 3 أشهر·discuss
> hot MILFs in my area

https://youtu.be/FBSam25u8O4
nojs
·قبل 3 أشهر·discuss
this is super cool, well done