Dear Philip E. Gibbs:
Quoting you, "If an artificial intelligence is able to interact with the universe and be aware of itself then it is certainly conscious." I fully endorse your view; consciousness is an emergent property. Nature allows intelligence and the best possible 'intelligence' evolves naturally and it has consciousness.
AI is our creation; we have to provide the environment for it; that is, AI evolves in an artificial environment. To 'interact with the natural universe' it will always need our help. To remove this middleman, we have to create an AI based on the structure nature has selected. That is, if ever humans create an 'AI' having consciousness, it will resemble humans; we will end up creating ourselves.
In my opinion, the theoretical existence of 'all possible universes' and 'all possible conscious beings' are identical and is the best premise to start with. However, at the end it may turn out that the laws of mathematics, together with top-down causal effects, restrict the possibility to just 'one unique structure for the universe' and 'one unique structure for the conscious beings'.
Jose P Koshy