In principle this makes sense; where the line around “your thoughts” is drawn is much more difficult to define in practice. What is a thought? Is it the words

In principle this makes sense; where the line around “your thoughts” is drawn is much more difficult to define in practice. What is a thought? Is it the words you hear inside your head as you think? The images? Do those even have a concrete biological representation? Are we just talking about patterns in neural signaling cascades? If so, I doubt that “thinking about a tree” looks the same from one person to the next. Which means that these patterns will have to be learned from training data (once the sensing technology exists). Brains exist to decide to take actions. In the above case, companies with an interface to your brain will only “understand” your thoughts insofar as they map to actions you can take through their services, or “API”, so to speak. In that sense, It feels like we’ve already crossed this line. Let me be more concrete: unless there were, say, a bomb manufacturer with a brain-API that you “trained” and used regularly, nobody will be able to decode that you think about bombs in your spare time. Unless, of course, a (hypothetical) brain computer interface just translates your thoughts into “stream of consciousness” strings of words - but if that were the case, how does that add value over a speech-to-text google search? TL;DR, if you’re worried about the thought police, look around you - we’re already there. Your “thoughts” are just the actions you decide to take using technology.