I think that visiting a drug rehab center would be much less convincing (though much faster) than the above suggested method. This is because a drug rehab center will look bad whether or not the effects are very rare, since it's selected for people who got bad enough effects to be in a rehab center.
(If his argument is that the bad effects don't exist, a rehab center would be good evidence against that, but it sounds like he believes more that they're rare and mild enough to be worth it.)
In general, if you want to convince someone who is taking ideas ...
I don't have much of an answer for you but wanted to explicitly thank you for posting this thread, I am in a similar situation and wouldn't have thought to ask here but should have.
[Hi! Been lurking for a long time, this seems like as good a reason as any to actually put something out there. Epistemic status: low confidence but it seems low risk high reward to try. not intended to be a full list, I do not have the expertise for that, I am just posting any ideas at all that I have and don't already see here. this probably already exists and I just don't know the name.]
1) input masking, basically for oracle/task-AI you ask the AI for a program that solves a slightly more general version of your problem and don't give ...
I may have seen this post too late for it to be of any interest, but I believe the Avalon Hill board game Magic Realm is just straightforwardly cohabitative - or at least, it defines an effectively cohabitative win condition, even if it also defines a competitive one. It's also unusually open/simulationist* for a board game, which seems likely to be correlated.
(* not in the sense of "realistic", but in the sense of "we have made a world and rules to determine what happens when you do something in it")
From the fan-written tutorial:
"At the end of the g... (read more)