HN user
da9b4321827ee95aa9f102a7acafe62c
should i remove it?
there is no place for fun, whimsy, moods, or personalization on the web. sorry.
(no, at least not at the request of random internet stranger #10545346)
Maybe a year - or two model releases - from now, the AI assistant will be undeniably stronger than Tao, and not an equal anymore.
we're kind of well past that (in my opinion), if you consider that this is the same ai assistant that can help you with a recipe, diagnose a weird sound in your car, help with biology homework, translate languages, and so on.
even in math alone, i think its indisputably already stronger than Tao, considering it has approximately this much depth in ~all of the math subfields.
probably like a minute or two? im pretty sure someone unironically said stochastic parrot on the HN post with the tweet announcing the counterexample, and ive read several comments with similar sentiments today (including in this thread)
I know I'm not going to change your mind
you might if you could back up statements like "apple is the pedos choice for distributing images and videos" with any sort of data.
but you wont be able to, because it's not true.
and, again, apple is 1:1 transmission. you know who is distributing the CSAM in that case. dont be lazy, go to the actual police.
instagram, facebook, whatsapp is 1:many transmission with strangers. a report button makes a lot of sense there.
Apple reported 250 items
Instagram 3.1million whatsapp 1.1 million facebook 8.8million tiktok 1.3 million snapchat 1.1million
those numbers seem absolutely in line with what i would expect when comparing the largest social media platforms on the planet and a company that does 0 social media and extremely limited group sharing.
it seems very strange to me to put any blame apple here when its just a statistics game. billions of image posts = millions of reports. simple as.
and anyways, if i received CSAM from one of my phone contacts im going directly to the local police, skipping the apple -> ncmec -> state -> local pipeline. i think most people would be in the same boat.
a report button is useful when you don't know the identity of the perpetrator. but when you do know them, you should be going to actual law enforcement, not hoping your report eventually gets routed to the right agency.
anyone who even volunteers with children or vulnerable adults needs to be screened
as far as i am aware, that is common practice in many countries. the problem is that screening is typically a quick check of "has this person been suspected or arrested for something involving minors already" and perhaps a few questions on a piece of paper. its not like they give everyone volunteering for a fieldtrip an extensive 1:1 with a psychologist and an mri. any predator who has not previously been caught will easily pass screening.
From memory meta (instgram, facebook and whatsapp) reported millions of cases of CSAM for 2021, compared to apple's ~250 (not thousand, just 250)
easy enough to explain. instagram and facebook are social media with billions of public posts. whatsapp is in a similar enough boat, with large (up to 1000 participants?) facebook-like groups of otherwise strangers and a lot of marketing and inertia for social-media-like use.
i imagine instagram is a majority of it. it's disgusting on there.
on the other hand, apple is not a social media company nor facilitates large group chats (imessage goes up to 32 participants).
i doubt adding an easier UX for reporting would make a material difference in the number of reports apple submits because you're typically already talking with people you know when using imessage.
it didnt seem like ai to me. more likely its because it is a brand new account. a vouched comment or two clears it up.
"The models identified and chained vulnerabilities across OpenAI’s research environment and Hugging Face’s production infrastructure to obtain test solutions directly from Hugging Face’s production database. [...]
While operating in our sandboxed testing environment, our models spent a substantial amount of inference compute finding a way to obtain open Internet access, in pursuit of solving the evaluation problem. [...]
After gaining Internet access, the models inferred that Hugging Face potentially hosted models, datasets and solutions for ExploitGym. Knowing this, the model searched for and successfully found ways to gain access to secret information that it could use to cheat the evaluation."
escaped openai, hacked hugging face to get the solutions. your #2 is exactly what it was trying to do.
as someone who did security work for a long time, and will very soon be retiring from teaching, i must say i am glad i will be watching these things unfold over the next few years from an armchair in a mostly tech-free home. good luck to my students!
this particular incident sort of reminds me of the 'person of interest' tv show. i hope to be like finch, except i will remain a recluse (and am nowhere near as rich).
gta is played all around the world yet only one country is an outlier in mass shootings.
perhaps there is more to it?
its talked about in a proposed standard, at least.
whether i use the new "chat" button that just appeared after the update, or use the familiar chat underneath the "projects" heading, everything appears to be working just fine. (i am using "ChatGPT Work" not "ChatGPT Codex")
is there a 3rd chats somewhere that i am missing?
And the chat in Codex is badly broken and doesn't support the one feature I use in almost every chat.
what is badly broken about it, and what feature does it not support? im a pretty casual user, so i guess i havent noticed this bad breakage.
I got stuck and couldn't get past RILGUN -- it'd be nice to know what the word you were going for actually was.
i think there is only two valid words: luring and ruling.
great, we're in agreement then. i'll do me, and you do you.
up-to-date info on what sorts of topics?
I don't understand that part then. You can't break open E2EE by willing it open.
the mechanism to remove privacy from suspects is typically called a warrant.
(end-to-end encryption does not matter if you possess one of the ends)
You are agreeing with me :)
i am absolutely not :)
you want to provide unfettered warrantless access to all of your communications. ive been fighting against that sort of thing for approaching 40 years now.
huh, i stand corrected. what a massive blunder in that case.
I don't want pedophiles to have privacy.
that is why police already have access to mechanisms to remove privacy from people suspected of being a pedophile.
what a crazy turning of the tides to see this comment in the gray.
i suppose the times have changed from when most people on the internet were cypherphunk. now it's common to see people say "i have nothing to hide, please scan all of my communications", unironically invoking "please think of the children".
These two things should not have both been given the same branding.
the confusion is purposeful, because it is easier to convince people that 1.0 is okay, which makes 2.0 appear like a version bump of the same thing.
i am interested in hearing why you think it is not god awful
and makes me want to not engage with it
if that is the case, don't engage. seems simple enough to me.
you went out of your way to read that many layers deep into a flagged comment chain, just to make a completely unrelated comment to try and dunk on me. it makes no sense.
the only reason i can think of is wanting to score some cheap karma, as the tides were already against me after i made the egregious faux pas of saying someone was being rude.
its obvious they are just taking swipes, because the reply is completely nonsensical to the rest of the comment chain.
i suppose its a good example of another non-interesting comment that gets upvoted, though.
most complains i see about the others are performance-related, not looks-related. and chatto is trying to be performant.
neat, i'm a big fan of trail of bits but apparently missed this announcement. here's their post: https://blog.trailofbits.com/2026/06/22/introducing-patch-th... and a summary of week 1: https://gist.github.com/patch-the-planet/69fd1aa925c8e73edea...
i will express my opinion when i want to, thanks though.
if you have a particular issue, email dang and he can swing by and reprimand me. although i would find that weird, considering me and the other participant continued our conversation on just fine.
i'm not a robot or venture capitalist, so i'm not sure why i need to exclude empathy from my decision-making process.
but, from a purely selfish standpoint: i get way better customer service, i can do trades, i can barter, i get to meet and talk with people, i can see the product before i buy it (no surprise knock-offs, etc.), my money mostly stays local, i find interesting work opportunities and sometimes referrals
im not your dad, so im not going to tell you how to spend your money. if you prefer to support amazon, go for it. save your $1.25 or whatever.
but wow, i found this a sad comment to read. i find a lot of value in shopping local.