Read in the same newspaper here: if an experimental approach to discovery is a crime, then we might as well put the Wright brothers, Charles Lindbergh and Apollo’s lunar-bound astronauts on trial Well, the difference is that those guys did not so recklessly endangered the life of others...

Think that should he have succeed with this submarine, he may have killed more innocent people in a bigger one, and who knows, some valuable ones. Therefore, a Darwin award may be in order.

...and curious social scientist will unplug it and watch what's happening

... so a big problem is that people may want to use it for anything instead of looking for a better solution. Stupids use stupid tools in stupid ways. No big surprise regarding the result then.

You can't have opt-in without opt-out, because the whole purpose of the opt-in option would be that you may opt-out later if you change your mind. So opt-out is the good strategy, and opt-in / opt-out come after (no opt-in without an opt-out option).

Let's them team up in jail, and see what kind of product they come up with.

Everybody should be able to choose to opt-out: as long as AI technologies are not totally safe (and explaining why an AI produces a given result is still far from being clear), people should be able to tell that they don't want to be considered as guinea pigs.

The PRC at his best: liars, thieves, revisionists, and soon-to-be aggressors...

Expect everything from the guys who don't want to ponder on how they could do better using a new tool.