So what happens when Mistral creates a collection of powerful AIs for different clients that have fundamentally different ethical frameworks from each other? Th

So what happens when Mistral creates a collection of powerful AIs for different clients that have fundamentally different ethical frameworks from each other? Then those AI interact? Do they politely agree to disagree? Or do they goal seek against each other?