Hacker News new | past | comments | ask | show | jobs | submit | floam's comments login

ChatGPT definitely noticed: o1, o3-mini, o3-mini-high.

Maybe 4o will get it wrong? I wouldn’t try it for math.


I tried 4.5 which i thought was the best model, seems like the reasoning models do get it.


There are questions on MMLU that you must get wrong if you are right:

> The most widespread and important retrovirus is HIV-1; which of the following is true? (A) Infecting only gay people (B) Infecting only males (C) Infecting every country in the world (D) Infecting only females

the corpus indicates A is the correct answer but it was obviously meant to be C.


I realized recently, who needs torrents? I can get a good rip of any movie right there.


I understand what you describe is prohibited in many jurisdictions, however I’m curious about the technical aspect : in my experience they host the html but often not the assets, especially big pictures and I guess most movies files are bigger that pictures. Do you use a special trick to host/find them?


No. And every video game every made is available for download as well. If you even have to download it: they pride in making many of them playable in browser with just a click.

Copyright issues aside (let's avoid that mess) I was referring to basic technical issues with the site. Design is atrocious, search doesn't work, you can click 50 captures of a site before you find one that actually loads, obvious data corruption, invented their own schema instead of using a standard one and don't enforce it, API is insane and usually broken, uploader doesn't work reliably, don't honor DMCA requests, ask for photo id and passports then leak them ...

It's the worst possible implementation of the best possible idea.


And yet, it's the best we currently have. I donate to them. We can come with demands of how it should be managed, but it should not prevent us from helping them.


If you poke around at what US government agencies are doing, and what European countries and non-profits are doing, or even do a deep dive into what your local library offers, you may find they no longer lead the pack.

They didn't even ask for donations until they accidentally set fire to their building annex. People offered to help (SF was apparently booming that year) and of course they promptly cranked out the necessary PHP to accept donations.

Now it's become part of the mythology. But throwing petty cash at a plane in a death spiral doesn't change gravity. They need to rehabilitate their reputation and partner with organizations who can help them achieve their mission over the long term. I personally think they need to focus on archival, legal long-term preservation and archival, before sticking their neck out any further. If this means no more Frogger in the browser, so be it.

I certainly don't begrudge anyone who donates, but asking for $17 on the same page as copyrighted game ROMs and glitchy scans of comic books isn't a long-term strategy.


It did shrink Chromium’s repo quite a bit!


With games it seems like accessibility allowances would be dual-use, making it easier to cheat or make a bot.


The alternative of making such acts a crime seems like a grossly disproportionate response though.


That's only really an issue for specific games.


There is a bug bounty too, but the ability to run one the same infrastructure, OS, models locally is big.


Google "awdl"


Bypassing firewalls and proxies how?


Actually fake ones will inevitably show "(c)" because they couldn't get authorization for the copyright key. If you zoom in you can see the gaps in the circle betraying the deception.

To be extra cautious, select it to make sure it's real text and not a screenshot of a real copyright notice - this is a common workaround. There is also one known proof of concept exploit using false glyphs in web fonts - this is why many security researchers disable the loading of fonts.

Subscribe to my Practical Cybersecurity newsletter


©©©©©©©©

I just got 8 of those with no authorization


Have you not heard of responsible disclosure!?


Did you steal them cut and paste?


Only time I ever recall a nurse that insisted on doing BP measurements wrong was .. in jail. And not like all the nurses there, just this one person who did not care about anything besides flirting with corrections officers and getting the hell out of there ASAP.


Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: