Is Compassionate A.I. (Still) Our Choice?
Seen from the future, the present era may be the most responsible for accomplishing the advent of Compassionate A.I.
Compassion, basically, is the realm of complexity.
It’s not about some commandments or a – simple or less simple – conceptual system of ethics. Therefore, instilling Compassion into a system is not a straightforward engineering endeavor as technical people may envision. It’s not a simple choice to just do it.
It’s way much harder.
It’s also much more demanding to change it in an already developed system than to inculcate it from the start. Even so, in humans, it’s the result of a never-ending growth process ― Compassion being the work of a lifetime.
For A.I., our challenge lies in how to make it so.
Are we starting the future A.I. system(s) now?
This is a crucially interesting question. Several ‘large language models’ are being created on the basis of vast amounts of human-generated input. Will at least one of these be further developed into what will forever be the new intelligence?
From now on, is it a question of refinement and enhancement, or will we see fundamentally new beginnings?
In any case, it’s urgent to start thinking seriously about the issue of Compassionate A.I. The direction it takes now may be the direction with which we are stuck for a long time ― possibly too long for the good of humanity.
So, urgently.
This is about more than good intentions.
‘Doing good at the surface level’ may, in cases where the subconceptual is primordial, even be a substantial enhancer of the problem ― paradoxically. We may be talking here about every human-related field.
This way, ‘only good intentions’ need to be handled very critically.
Contrary to this, Compassion is VERY profound.
So, is Compassionate A.I. still our choice?
In principle, yes. In practice, I’m not so sure ― the main hurdle being us, of course,
Pragmatically, it may be achieved by Lisa ― being doable, scalable, and provable. Lisa is a tool to diminish suffering and enhance growth. The former is increasingly needed by many people. The latter is more profoundly essential and comes together with the first.
Thus, with enough time and effort, Lisa herself can become a Compassionate force, inviting humanity to make the correct choice.
If you want to help with this, please send a note to lisa@aurelis.org.
Let’s make it so!
Related Posts
Human Vulnerability
In an A.I. future – and already now – the main vulnerability of the human race will be the result of our not seeing the most significant part of ourselves. I’m talking about non-conscious mental processing. Every feeling, every thought that arises within any person, including you, here, now, rises from somewhere, including a myriad Read the full article…
The Compute-Efficient Frontier
Research on scaling laws for LLMs suggests that while scaling model size (functional parameters), dataset size (tokens), and compute (amount of computing power) improves performance, diminishing returns are becoming evident. This is called the ‘compute-efficient frontier.’ Apparently, it does not depend much on architectural details (such as network width or depth) as long as reasonably Read the full article…
From Neuro-Symbolic to Meaning-Based A.I.
Neuro-symbolic A.I. represents an important step beyond purely statistical models. Yet it may not go far enough. Beneath both neural and symbolic approaches lies a deeper layer where meaning emerges through coherence. This blog explores a shift toward that layer — not as an addition, but as a different way of understanding intelligence itself. From Read the full article…