Doubt it, there was a streamer doing minecraft experiments with all the different SOTA models and pitting them against eachother in a survivor-esque style reality TV show competition and they had quite a hard time.
Turing Games. It's hilarious. Although it owes a lot of the entertainment value to the human touches, such as giving the AIs different TTS voices (Llama 4 has a Texas dialect!) and letting them evolve their characters from game to game.
Honestly when we get long session histories or continuous learning it's going to get double wild. It's another experience when it can reference a years long history that is individual to you working with it.