<!-- bryant-content-generator:start --> # Safety systems in conversational AI may govern the... > Safety systems in conversational AI may govern the urgency, tone, and care of a response, but they must not govern whether evidence is classified as support, contradiction, or uncertainty when the system evaluates descriptions of reality against what is known to exist. **Source:** [[articles/Epistemic Invariance in Conversational AI|Epistemic Invariance in Conversational AI]] ## Share on Social Media ``` “Safety systems in conversational AI may govern the urgency, tone, and care of a response, but they must not govern whether evidence is classified as support, contradiction, or uncertainty when the system evaluates descriptions of reality against what is known to exist.” — Bryant McGill https://bryantmcgill.com/quotes/quotes-safety-systems-conversational-against-known-exist ``` <!-- bryant-content-generator:end -->