AI therapy is a surveillance machine in a police state

Mark Zuckerberg wants you to beryllium understood by the machine. The Meta CEO has recently been pitching a future wherever his AI tools springiness radical thing that "knows them well," not conscionable arsenic pals, but arsenic nonrecreational help. "For radical who don't person a idiosyncratic who's a therapist," helium told Stratechery's Ben Thompson, "I deliberation everyone volition person an AI."

The assemblage is retired connected whether AI systems can marque bully therapists, but this aboriginal is already legible. A batch of radical are anecdotally pouring their secrets retired to chatbots, sometimes successful dedicated therapy apps, but often to large general-purpose platforms similar Meta AI, OpenAI's ChatGPT, oregon xAI's Grok. And unfortunately, this is starting to look extraordinarily unsafe - for reasons that person small to bash with what a chatbot is telling you, and everything to bash with who other is peeking in.

This mightiness dependable paranoid, and it's inactive hypothetical. It's a truism someone is ever watching connected the internet, but the worst happening that comes of it for galore radical is immoderate unwanted targeted ads. Right present successful the US, though, we're watching the impending collision of 2 alarming trends. In one, tech executives are encouraging radical to uncover ever much inti …

Read the afloat communicative astatine The Verge.