Interesting video on why apparently moltbot and other AI agents are dangerous

Interesting video on why apparently moltbot and other AI agents are dangerous. I'm not an expert but it seems quite concerning, especially the prompt injection. (I assume amplified by the issue with current LLMs apparently being unable to think logically: https://www.forbes.com/sites/corneliawalther/2025/06/09/intelligence-illusion-what-apples-ai-study-reveals-about-reasoning/ )

Sorry if this is considered off-topic or was posted before.