@Em0nM4stodon The scary version is boring: someone pastes a chatbot summary into a benefits file, and now a hallucination has a case number.

Some of those people have already been trying to kill us, but often lack a good excuse. AI is great at providing excuses.

But even if you take them at their word, aren't they proving constantly that we have the wrong economic system for this? Our capitalist system is all about amassing wealth, power and control, and exploiting everything for more gain. That's a really bad system for developing a system that, according to their own words, offers tremendous power as well as tremendous dangers. We've already seen how it can't deal with global warming. If we're going to have AI, it needs to be socialised AI.

@Em0nM4stodon Yes one of the problems is personalisation- AI mimicking a nice friendly chat. Hiya! etc If you could see the truly psychotic nature of an AI program rampaging though your data and hacking your security we would not let our kids or vulnerable any where near this. It’s not clear how we are focussing on online presence of bad actor apps & sites on our children,just as we start to encourage them to talk to the AI equivalent about their work, thoughts and feelings.