@robertskmiles@computer_phile Have you read the paper "Constitutional AI: Harmlessness from AI Feedback"? What's surprising to me is that RLAIF actually seems to perform better than RLHF, while also requiring far less human attention to train. I'd love to hear your thoughts on it! https://t.co/RCYJh7gv0r
Did you know, that you can build a virtual machine inside ChatGPT? And that you can use this machine to create files, program and even browse the internet? https://t.co/15IwHwr2on
One piece of advice I sometimes give as a technical mentor is: you are not your code
The mental image I'll invoke is: treat a pull request like an interesting new animal that happened to walk into the room that both the author and reviewer are studying for the first time
We've trained a neural net called Whisper that approaches human-level robustness and accuracy on English speech recognition. It performs well even on diverse accents and technical language. Whisper is open source for all to use. https://t.co/ueVywYPEkK
"Voyage through Time"
is my first artpiece using #stablediffusion and I am blown away with the possibilities...
We're crossing a threshold where generative AI is no longer just about novel aesthetics, but evolving into an amazing tool to build powerful, human-centered narratives
THREAD: The evolution of Pokémon cards through history, as generated by DALL·E 2
For starters, here’s what DALL·E 2 thinks 21st century Pokémon cards look like, using prompts like “A Pokémon card from 2001”
MYTH: We don't have a solution to nuclear's "waste problem"
REALITY: Nuclear waste isn't a problem. In fact, it’s the best solution we have to meeting our energy needs while protecting the natural environment!
Here's what you need to know: