I also this that these conceptual models are probably the biggest reason AI/LLMs/ML will be a big problem. People aren’t wired to defend against these tools.