cfgt retweeted
We all know ChatGPT can be jailbroken. But what makes these attacks possible?
In a new paper (w/ @nhaghtal @JacobSteinhardt), we delve into how safety training fails, identifying two key failure modes that enable jailbreaks: competing objectives and mismatched generalization
1/
cfgt retweeted
Times ChatGPT hallucinated functions existing in the rust ffmpeg bindings: 152
Times it got the code right: 0
For strange shit, we just aren’t there yet. But if you are doing something non-complex sure you might be able to tab complete your way there ¯\_(ツ)_/¯
cfgt retweeted
Nothing like seeing people complain about something you have in github open for PR's without actually making any effort to fix things.
cfgt retweeted
AWS keynote Sydney. “Data beats intuition”. Really? If data defines everything at Amazon why is the search ranking so crap. #awssummitsydney
cfgt retweeted
Reinforcement Learning by Human Feedback is just parenting for a supernaturally precocious child.
cfgt retweeted
page 37 of the GPT-4 paper: cdn.openai.com/papers/gpt-4.…
cfgt retweeted
This is absurd. To check if I'm a robot, they're using AI generated images and none of them look remotely like basketballs.
Yet *I'm* the one who's wrong.
cfgt retweeted
Cones: Concept Neurons in Diffusion Models for Customized Generation
abs: arxiv.org/abs/2303.05125
cfgt retweeted
A huge credit to the make-up artist. She previously worked on Mad Max where she was usually applying black eyes rather than hiding them.
well. i've recently had a lot more free time than usual, and…
… I should have taken more breaks and paced myself, but …
touchhle.org/
yeah. enjoy!
also, a huge thanks to my friends cassie, erin, puck and mary for help at various points.
cfgt retweeted
I've seen a lot of people asking "why does everyone think Twitter is doomed?"
As an SRE and sysadmin with 10+ years of industry experience, I wanted to write up a few scenarios that are real threats to the integrity of the bird site over the coming weeks.
cfgt retweeted
in this thread: man spends an entire day trying to get a bloody ceiling fan to go on the internet ☠️
Losing my mind a bit with the @bigassfans app. Connect to their broadcast network, finds the device, join it to my network then… nothing. 3 different fans, exactly the same experience. 1/n
cfgt retweeted
they call it "single sign on" but they should call it "dozen-flickering-redirects sign on"
cfgt retweeted
My son has lived through four chancellors, three home secretaries, two prime ministers and two monarchs.
He's four months old.

































