"Companies like OpenAI must develop technical safeguards, he argues, but also points out that it is difficult to effectively combat misuse without a clear legal framework and public debate"
In others words, always listening, always tracking, and always violating you - all in the name of "saving you from yourself", of course. 😏
Another thing, Sam is either lying or he is really lost in his own system today; because anyone could test this out for themselves, that these Ai systems are already heavy political bias with its answers... today, not tomorrow or when the "next ChatGPT 5.0+ is released.
If anyone that is interested in sampling these Ai ecosystem very heavy political bias answers, HONESTLY, engage with Copilot, ChatGPT, ect. and ask the Ai system something from an all points of views perspective on a controversial topic (there's oddles of them today, so, just pick one)... and just absorb which direction the Ai assistant will heavily lean towards its answer.