The slow collapse of critical thinking in OSINT due to AI

It's become almost comical to me to read articles like this and wait for the part that, in this example, comes pretty close to the beginning: "This isn’t a rant against AI."

It's not? Why not? It's a "wake-up call", it's a "warning shot", but heaven forbid it's a rant against AI.

To me it's like someone listing off deaths from fentanyl, how it's destroyed families, ruined lives, but then tossing in a disclaimer that "this isn't a rant against fentanyl". In my view, the ways that people use and are drawn into AI usage has all the hallmarks of a spiral into drug addiction. There may be safe ways to use drugs but "distribute them for free to everyone on the internet" is not among them.

It’s a rant against the wrong usage of a tool not the tool as such.

It's a tool that promotes incorrect usage though, and that is an inherent problem. All of these companies are selling AI as a tool to do work for you, and the AI _sounds confident_ not matter what it spits out.

> All of these companies are selling AI as a tool to do work for you, and the AI _sounds confident_ not matter what it spits out.

If your LLM + pre-prompt setup sounds confident with every response, something is probably wrong; it doesn't have to be that way. It isn't for me. I haven't collected statistics, but I often get decent nuance back from Claude.

Think more about what you're doing and experiment. Try different pre-prompts. Try different conversation styles.

This is not dismissing the tendency for overconfidence, sycophancy, and more. I'm just sharing some mitigations.