3 ms·
ChatGPT answering the first would be much more embarassing for OpenAI than ChatGPT answering the second.
by sebzim4500 2y ago
ChatGPT answering the first would be much more embarassing for OpenAI than ChatGPT answering the second.
- option 2y agobingo
- ilikehurdles 2y agoWhen you realize “safety” applies to brand safety and not human safety, the motivation behind model lobotomies make sense.
- renewiltord 2y agoThat's what people care about, too. For instance, most people would rather have many hit and run drivers than have one autotaxi hurt someone.