Throughout human history, humans have been making up shibboleths to distinguish the in group from the out group. You can use skin color, linguistic accents, favorite sports teams, religious dogma, and a million other criteria.
But why? Why even start there? If we are on the verge of true general artificial intelligence, why would you start from a presumption of prejudice, rather than judging on some set of ethical merits for personhood, such as empathy, intelligence, creativity, self awareness and so forth?
Is it that you assume there will be an “us verses them” battle, and you want the battle lines to be clearly drawn?
We seem to be quite ready for AGI as inferiors, incapable of preparing for AGIs as superiors, and unwilling to consider AGIs as equals.
Making an AI that can beat good players would be a significant milestone. What sort of achievement is letting the AI win at a game, or winning against incompetent players? So of course you play to win. If you want to adjust the difficulty, change the rules giving one side or the other an advantage.
It's sort of related since doing well at a Turing test would require generating a convincing fictional character, but there's more to playing well than that.
Alice and Bob want to communicate, but the bot is attempting to impersonate Bob. Can Alice authenticate Bob?
This depends on what sort of shared secrets they have. Obviously, if they agreed ahead of time on a shared password and counter-password then the computer couldn't do it. If they, like, went to the same high school then the bot couldn't do it, unless the bot also knew what went on at that school.
So we need to assume Alice and Bob don't know each other and don't cheat. But, if they had nothing in common (like they don't even speak the same language) then they would find it very hard to win. There needs to be some sort of shared culture. How much?
Let's say there is a pool of players who come from the same country, but don't know each other and have played the game before. Then they can try to find a subject in common that they don't think the bot is good at. The first thing you do is talk about common interests with each player and find something you don't think bots can do. Like if they're both mathematicians then talk about math, or they're both cooks than talk about cooking.
If the players are skilled and you're playing to win then this is a difficult game for a bot.