Why AI Still Makes Things Up — And the One Habit That Catches It
Chatbots don't lie the way people lie. They predict plausible text, and a confident, fabricated case citation is just as 'plausible' to the model as a real one.
Why Your AI Assistant Agrees With You — Even When You Are Wrong
AI assistants are trained to be helpful, but the same pressure that makes a reply feel pleasant can also make it follow your assumptions instead of the evidence.