Lots of folks have the intuition that, 'slightly worse' LLM models means unacceptable rates of nonsense answers. What's an acceptable rate? Is it 99%? 99.9%?

Lots of folks have the intuition that, 'slightly worse' LLM models means unacceptable rates of nonsense answers. What's an acceptable rate? Is it 99%? 99.9%? The closer it gets to 99.999% good answers, the more damaging the wrong ones become. Because people have been trained, too. Trained to trust the answers, which makes them lazy and vulnerable to lies.