#ChatGPT Session Review - In our first targeted test, we explore the nature of additional censoring algorithms built in to GPT3 which can override language and censor certain topics. We discuss how these editing mechanisms work within our own minds, and how these concepts should be treated in the future of AGI.
A very timely article posted by Open AI that touches on many of the topics discussed here!
https://openai.com/blog/how-should-ai-systems-behave/