Entrusting our speech to multiple different corporate actors is always risky. Yet given how most of the internet is currently structured, our online expression largely depends on a set of private companies ranging from our direct Internet service providers and platforms, to upstream ISPs (sometimes...
“ISPs should monitor and censor bad things!”
“ISPs shouldn’t be censoring free speech!”
We already see how bad stuff gets censored or punished on social media with things like bots that cast such a wide net that lots of innocents get caught with no human to appeal to.
You only have to look at how the first amendment is abused in the US to understand just how bad free speech without consequence is.
Yuval Noah Harari did a session with the rest is politics podcast. He brought up a concept that was total alien to me beforehand. In the past fake news has sought to shock people into taking notice. With AI this will change things dramatically. Rogue states will use AI to befriend individuals, and then manipulate the thought process by gentle integration. I found this an immensely scary prospect to dwell on. People rarely think about the person on the other end of a conversation being something other than what they portray. The concept is very credible.
To me, this is one of the reasons why we must police activity online. But it must be done without government interference. Ideally an international effort should be made. An international fact authentication group would go along way also.
Edited for better clarity.