Can businesses trust AI?

Why does artificial intelligence sometimes make things up? From law firms to consultants, businesses are grappling with the risks of AI hallucinations. What's causing them, and can they be prevented?