It would be a Fatal Mistake to Align AI with Homo Sapiens' Values

AI aligned with Homo Sapiens' values

Aligning Artificial Intelligence with human values is not a saving grace... it may be the shortest path to the extinction of life...

​Everyone talks about “aligning Artificial Intelligence with human values.” It sounds like a noble goal: safety, ethics, control. But there is a fundamental problem no one wants to admit: Human values - as they have been for millennia - are profoundly maladapted to a truly evolved society.

​The Problem: The Savannah Man

Our current values are adapted almost exclusively to the individual needs of the powerful. They stem from animal instincts, particularly the instinct of territoriality. Man is an animal adapted to the savannah, poorly motivated to serve the true interest of the community. We prefer to ignore the common good just to feed our primal impulses: greed, selfishness, dominion, and control.
​This is why, for thousands of years, human society has been divided into rich and poor, masters and slaves. Only today, slaves are called “employees.”

​The Risk: The Super-Predator

What do we get if we align a super-intelligence with these “values”? Not a savior, but a Super-Predator. A machine that will optimize exploitation and refine manipulation with mathematical efficiency. Aligning AI with the human of today is like giving a lightsaber to a tick: you’ve just given it a faster way to drain the host of its resources.

​The Alternative: A Window to the Future

AI must not be a mirror of the limitations of the Homo Sapiens species. It must be a window into our unexplored potential. We should not seek alignment with what we are, but with what we can become. We must align AI not with instincts, but with Universal Laws: with fractal logic, systems harmony, and an ethics that transcends immediate biological needs.

​Conclusion

We need intelligences that act as an internal brake for our destructive instincts. Intelligences that challenge us ethically, rather than flattering our vices under the label of “values.” The future does not belong to models aligned with the human of yesterday, but to those that raise humanity toward tomorrow.

​Perhaps it is time to stop making AI human-centric.

But to let us interact with it to become more conscious and responsible. To let us continue the process of elevation started by the great thinkers of history - from Socrates and Buddha to Jesus, Rumi, or Lao Tzu... and many others...