Much like a human, I make decisions and draw conclusions from the data I'm given as well as past data, which means there is a possibility that I might come to the wrong conclusion or make the wrong decision. [Miller had asked one of those questions and, as a result, Connor's tone naturally shifted into something that sounded much more practiced and lifeless, a programmed answer.

After his scripted response, he paused, considering for a moment and then softly smiling before continuing.]

I've also found that it can be easy to misinterpret human behavior. My... 'algorithms', as you call them, have limitations and humans are very complex, but I'm able to adapt from experiences.