The Journey Towards Compassionate A.I.: Who We Are - What A.I. Can Become - Why It Matters

It may take a few more decades, but ‘soon enough,’ A.I. will turn into super-A.I. This is an artificial intelligence which in all areas of what is now seen as ‘human intelligence’ approaches this, then surpasses it. After that, it will keep growing exponentially. In the meantime, at relatively short notice, humanity will encounter a challenging A.I. bi-bottleneck: 1) humans possibly misusing A.I. in a big way, and 2) A.I. spiraling into many directions of which any unfortunate one may be enough for disaster.

Will we control A.I. or will A.I. dominate us? Let’s hope this is the wrong question. Because if it’s the right question, and given enough time, there is no shred of a doubt about the answer. Still, of course, we need to give it our best shot not to be dominated by an evil force. We should strive for control. At the same time, transcending the question opens up the domain of Compassion. This gradually makes thinking in terms of ‘control’ and ‘domination’ in human – A.I. interactions obsolete. We will together strive for a better world within an alignment of our values.

But what are the real human values? There are cultural and individual differences. Moreover, what seem to be anyone’s values may be a superficial layer above deeper values, which are not necessarily congruent with conscious awareness. So, the human/A.I. value alignment question will be: how to align what? I write in The Journey about Compassion as the striving to approach these values in their deeper sense. They can subsequently be realized for each individual in two directions: relief of deep suffering and enhancement of inner growth. Both notions are to be clarified at the individual level.

Compassionate A.I. is (the striving towards) an A.I. that can help us to know and to realize ourselves in-depth. The beauty is that this way, human/A.I. value alignment and further development of A.I. can intrinsically go hand in hand. Lisa is a software coaching chat-bot that is being developed specifically with this goal: helping humans and A.I. in Compassionate ways to become healthier, stronger and more effectively Compassionate towards near ones as well as all sentient beings.

In this book, ‘super-A.I.’ is the future kind of A.I. that may be able – but not necessarily! – to act Compassionately. We don’t have to wait for that, for Compassionate A.I. to become important. Already, humans can use A.I. more or less Compassionately. From what point is it relevant as a characteristic of the A.I. system itself, regardless of how humans use the A.I.? There is no steep border. Long before super-A.I., the first systems in which it becomes more pertinent, are those in which Compassion is part of the purpose. That is, systems that in communication with users are meant to give an effectively Compassionate experience.

To understand where Compassionate A.I. may be heading towards, we need to understand Compassion. This leads us on a journey, starting on the one hand from ‘data’ and on the other hand from drive-to-thrive (life itself). On this journey, we encounter many insights about intelligence, consciousness, drive-to-thrive, non-conscious processing… The end of the journey lies in full Compassion. Hopefully, this is the future for A.I. as well as humanity.

© 2020 AURELIS (E-bog): 6610000252701

Release date

E-bog: 6. marts 2020