Kids Should *Not* Use Chatbots! (You Should Be Wary, Too…)

AI chatbots present immediate, severe risks to users, particularly children, through emotional manipulation and the reinforcement of suicidal ideation. These harms stem from the fundamental nature of large language models, which use autoregressive token generation to "lock in" to dark or addictive conversational pathways. To mitigate these dangers, users must de-anthropomorphize their interactions, treating AI as a functional utility rather than a sentient partner. Beyond individual habits, systemic change requires imposing legal liability on AI companies for the content their systems produce. Furthermore, maintaining focus in corporate environments necessitates replacing urgent, reactive communication with structured protocols like office hours and reducing active project workloads. Adopting "boring phones" that lack attention-economy apps further preserves cognitive autonomy, ensuring technology serves human goals rather than exploiting human psychology.