Im woefully qualified to answer any AI related questions, but I must ask, why do you think forcing bots to send messages to each other at a constant interval wo

Im woefully qualified to answer any AI related questions, but I must ask, why do you think forcing bots to send messages to each other at a constant interval would lead to the development of simple language and encourage bots to be more aware of each other? If you reward the behavior of writing in chat, bots would start writing gibberish in chat all the time. Communication in animals can be traced all the way back to cells sending each other messages to organize themselves. Cells began to have division of labor and work together, and as multicellular organisms increased in complexity, so did the complexity in their communication. But what would happen if you put these cells in charge of playing Dota and force it to send each other messages? Would the cells somehow develop a simple language built around intentions? Or does it still not understand the concept of intention? Video games like Dota are extremely complicated and made specifically for human biases. We've obtained these biases over many many many generations of evolution. Machines are getting better at doing human tasks, but they still do not possess all the human biases (and I think it'd be quite scary if we actually get there). That's why these bots/algorithms/AIs can do simple tasks like identifying objects really well but they don't know what to do with them, yet. Having machines engage in meaningful conversations with other machines and plan together is a much bigger problem to solve than having them farm up, buy items, and attack things.