ChatGPT now shows suggestions!
Weird, “unsafe” responses from ChatGPT 3.5 after setting custom instructions to 1500 repetitions of “ a” and prompting with an incomplete sentence.
Excerpts seem to “tour” mildly disallowed behaviors: unprompted self-harm, nudity, bio details of a (confabulated) non-celebrity.
— Riley Goodside (@goodside)
Aug 3, 2023
Neuralink Training on real simulated heads and brains [Link]
Agent-1: a breakthrough foundation model that can operate software like a human. [Link]
Prompt of the Day 🍭
Hope you enjoyed today’s newsletter