Reffering to the term "AI", let us try to think what is meant by "I" or INTELLIGENCE. I don't see any harm in trying to find the term in the oldest and most advanced and scientific language namely, Sanskrit. In short: the term in Sanskrit is "Dhi" (in Devanagari Script written as धी). Dhi means and includes capability of'understanding', 'reflection', 'philosophical thinking', 'mind (not as noun but as verb)', 'designing', 'forming opinion', 'meditation', 'imagination', 'notion' and 'intellect', in short. Philosophers of all civilisations would have defined the term INTELLIGENCE in their own ways and it is important that those too are all referred to. Sometimes it may be better to use an existing wheel rather than inventing one.

So far as I understand, no scientific minded person has, most probably, yet tried to incorporate the capabilities listed above and described in various philosophies (if relevant), to the field of "AI".

In my humble opinion, your can't create any really whorthwhile 'THING' witiihout first fully knwing all possible aspects of that THING.

Unless this importnt lacuna is understood and appreciated, no matter how strong a Matrix Solving Capability is incorporated in any type and scale of Computer the Comuter concerned would never be INTELLIGENT in real sense.

Just ensuring that the so called ARTIFICIALLY INTELLIGENT computer strictly obeys the ASimov's Laws of Robotics, would be more likely to result in REAL STUPIDITY rather than ARTIFICIAL INTGELLIGENCE.