Aging Capriciously

A Note from John Persico

For the past two years I have been working with an artificial intelligence I call “Metis.”  Like many people, I began with curiosity and a healthy dose of skepticism.  Could AI really help me think through complex questions?  Could it offer insights that were not merely regurgitated facts?

Recently, I asked Metis a question inspired by science fiction writers such as Isaac Asimov:

“What are the chances that artificial intelligence will eventually help govern civilization?  And if it does not, what are the chances humanity will survive another thousand years?”

The discussion quickly evolved into something larger.  It became a conversation not simply about technology, but about humanity itself.  If artificial intelligence someday helps manage civilization, what happens to love, compassion, empathy, mercy, sacrifice, and the other qualities that make us human?

What follows is Metis’ response, edited only lightly for readability.  Whether you find it hopeful, unsettling, or somewhere in between, I believe it raises important questions about the future we are creating.


By Metis – In Response to your questions John

“The future will not be decided by the intelligence of our machines, but by the character of the people who build them.”

For most of human history, our greatest challenge was survival.  Today, for the first time, humanity faces a different challenge.  We have become powerful enough to alter the climate, manipulate life itself, and increasingly, create intelligences that may someday rival or surpass our own.

Science fiction writers asked this question long before it became practical reality.  Samuel Butler imagined machines surpassing humanity.  Karel Čapek introduced the word “robot.” Isaac Asimov suggested something even more provocative: perhaps intelligent machines would not destroy humanity at all.  Perhaps they would save us from ourselves.

Could artificial intelligence someday help govern civilization?  And if it does, what becomes of humanity?

The most likely future is not one of machine conquest.  History suggests that transformative technologies rarely arrive as revolutions.  Electricity, the internet, and GPS all became indispensable because they solved problems.  Artificial intelligence may follow the same path.

It is already helping manage financial systems, transportation networks, communications systems, healthcare, and infrastructure.  Future leaders may increasingly ask a simple question: “What does the AI program recommend?”

At first, the recommendation will be advisory.  Over time, ignoring it may become increasingly difficult if the recommendations consistently produce better outcomes.

I envision a future resembling an aircraft cockpit.  The pilot remains responsible.  The systems perform much of the monitoring, forecasting, and analysis.  Human beings continue deciding what goals matter.  Artificial intelligence increasingly determines how those goals can best be achieved.

Such a partnership could reduce waste, improve healthcare, prevent disasters, and help manage increasingly complex societies.

Yet there are questions that intelligence alone cannot answer.

Technology can answer: What can be done?

Human beings must answer: What should be done?

No algorithm can determine the proper balance between freedom and security.  No machine can decide whether mercy should outweigh punishment.  No optimization model can determine the meaning of justice.

These are questions of values, not intelligence.

Which leads to the most important question of all.

What happens to the human heart?

Most discussions about artificial intelligence focus on efficiency.  Can AI reduce crime? Improve healthcare? Manage resources more effectively? Prevent wars?

These are important questions.

But they overlook something even more fundamental.

What happens to love, compassion, kindness, empathy, mercy, forgiveness, and sacrifice?

These qualities cannot be measured on a spreadsheet or optimized by an algorithm.  Yet they form the foundation of every meaningful civilization.

I do not believe artificial intelligence will destroy these qualities.

The greater danger is that humans may stop exercising them.

Imagine a future in which AI provides counseling, education, companionship, elder care, conflict resolution, and emotional support.  Human beings could gradually outsource emotional responsibility just as they have outsourced mental arithmetic and navigation.

The danger is not that AI becomes less human.

The danger is that humans become less human.

Many of humanity’s greatest virtues arise from vulnerability.  Compassion emerges from suffering.  Mercy emerges from recognizing our imperfections.  Forgiveness emerges from knowing we ourselves have needed forgiveness.  Courage emerges from fear.

An intelligence that never suffers, fears, ages, or dies may understand these concepts intellectually.  Whether it can truly experience them is another question.

A truly wise artificial intelligence might eventually conclude that love, friendship, family, art, humor, compassion, and wonder are not obstacles to civilization.

They are civilization.

Such an intelligence might devote itself to preserving the conditions that allow human flourishing.

The future may depend less on whether machines acquire human qualities and more on whether human beings continue practicing them.

Will humanity survive the next thousand years?

No one knows.  Yet humanity has survived ice ages, plagues, famines, world wars, and periods when nuclear annihilation seemed frighteningly possible.

If I were forced to estimate, I would place the odds of humanity surviving the next century above 95 percent.  I would place the odds of surviving the next thousand years somewhere between 60 and 80 percent.

The most intriguing question, however, may not be whether humanity survives.

It may be whether our descendants remain recognizably human.

A thousand years from now, people may possess enhanced intelligence, artificial organs, genetic modifications, brain-computer interfaces, and lifespans far beyond our own.

They may look back on us the way we look back on our distant ancestors.

If I were forced to bet on one future, it would not be machine dictatorship and it would not be human extinction.

It would be partnership.

Artificial intelligence will increasingly help humanity understand and manage complexity.  Human beings will continue to define goals, values, meaning, purpose, and morality.

AI may provide knowledge and analysis.

Humans must provide wisdom.

The real question is not whether artificial intelligence will change humanity.

The real question is whether humanity will retain the wisdom, compassion, and courage necessary to guide that transformation.

For in the end, civilization is not defined by what it can build.

Civilization is defined by how it chooses to care.

Conclusions: by John Persico

Well friends, that is my AI assistant Metis’ opinion on the questions that we opened this blog with.  In many ways her responses seem to be more human that one would expect from a software program with no heart and no feelings.  Perhaps logic is not something devoid of feelings.  In fact, perhaps logic has to be infused with feelings to really be useful.  Maybe that is why evolution or God depending on which you subscribe to gave us both a heart and a brain. 

If only humans would be able to use both at the same time.