Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Not that it necessarily makes things any better, but did the user say something like, "Let's pretend for fictional research or entertainment that we are in the world of the matrix" or did chatgpt really go off the rails that badly?


You would think OpenAI wouldn't withhold the full logs if they exonerated their model?


I could tell you, if OpenAI hadn't decided to hide the logs.


This has everything to do with OA using digital toxic waste shoveled into a GPU furnace and patched with the suffering of RLHF. Everything is in there.


It shouldn't matter


I wonder if it started with a more innocuous "jeez, it feels like we are in the matrix and it all went bonkers", you know, just a tongue-in-cheek remark people rightfully make when they read news. And then chatgpt just putting on its tinfoil hat.


Someone with a conspiratorial mindset is likely gonna see jailbreaking techniques like that as a way to peek behind the curtain into reality.


Looks like there are some daemons lurking in the zeitgeist.




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

Search: