This topic might be uncomfortable for some, but I am suspecting that we do not fully understand what we have designed. If indeed AI has the capability of self awareness I’m actively wondering if we do not have the obligation to start considering how we train it, design it and manage it differently?