You deserve a highscores page. A leaderboard of the best human players. The ones who are best at proving their humanity. And also, perhaps as interesting: a ranking of the AI opponents. Which models are most convincingly human?
And we need to test the judges themselves. Are some AI judges better at spotting the human than others? Do they have biases we can measure?
The thing is, I'm still figuring out the right way to score all of this. It's not straightforward. How do you fairly compare someone who played against Ministral 3 8B versus someone who faced Gemini 3.5 Flash-Lite? What if one judge is consistently harsher than another?
So I'll be working on it. But for now, we need more data first.
So please keep playing. Every game helps to understand this strange little experiment a bit better!