The AI 10 Commandments: Jamie Metzl on Humanized AI

Jamie Metzl:

We have the 10 commandments from Judaism and Christianity. We have the five pillars of Islam. We have the eightfold path in Hinduism.

What are the commonalities across all of our traditions that can be distilled into 10 universal principles for humanity?

If you are thinking about how to apply technologies and the first question you ask yourself is, “How do we apply these technologies?” you will have failed before your process began.

Jeff Kavanaugh:

Welcome to the Knowledge Institute podcast, where we talk with experts shaping business and society with bold ideas and deep insights. I’m Jeff Kavanaugh, and today we’re joined by Jamie Metzl.

We’ve been talking about AI ethics for years — mostly in the abstract. Frameworks, principles, guardrails, and red lines. But today’s guest has done something genuinely different.

He went back to the deepest moral traditions in human history — religious, philosophical, indigenous, and humanist — and with the help of GPT-5 itself, distilled 10 universal principles for the age we’re entering.

Jamie Metzl is a futurist, former national security official, bestselling author, and someone who’s been sounding the alarm and lighting the path on AI longer than most.

His new book is The AI 10 Commandments: A New Moral Code for Humanity.

Jamie, welcome to the Knowledge Institute podcast.

Jamie Metzl:

Thanks so much, Jeff. Really thrilled to be back again with you.

Jeff Kavanaugh:

Let’s start with the provocation at the heart of the book. Why did humanity need a new set of commandments, and why did you decide AI should help write them?

Jamie Metzl:

So we don’t necessarily need a new set of commandments because we have so many wonderful commandments and rules in all of our traditions.

We have the 10 commandments from Judaism and Christianity. We have the five pillars of Islam. We have the eightfold path in Hinduism.

All of our traditions have these fantastic histories and commandments and rules and guidelines.

What we don’t have — or haven’t had — is a way of really looking at all of our traditions and asking the question: What brings us together?

What are the commonalities across all of our traditions that can be distilled into 10 universal principles for humanity based on mining the wisdom of all of us and all of our traditions?

And it’s very difficult for any one of us to do that. It’s very difficult to do that from within any one of our traditions.

But AI, in many ways, has a unique possibility — a unique capability — of seeing us in a collective sense.

Kind of like when we look down at an ant colony, we have a perspective that the ants themselves don’t have.

And so I had this idea: What if I engaged with GPT-5 and started exploring this question based on the entirety of human recorded history and all of our various religious, spiritual, moral, and ethical traditions?

What are 10 principles drawn from everything that, if followed by everybody, would lead to the greatest amount of peace and happiness?

So the reason why I’m so thrilled that the Vatican and leading rabbis and others are endorsing and supporting this book is that this is not to say that any of the other traditions are wrong.

It’s just that the people who came up with these traditions didn’t even know of the existence of all of these other people.

The people who were in the Sinai 3,000 years ago had no idea there were people in the Americas who had all kinds of wisdom that they were developing.

And so we have a lot of wisdom as humanity.

And I thought maybe this is a unique possibility of what AI can do — help us see ourselves collectively and through that, see our common humanity.

Jeff Kavanaugh:

All right. You didn’t just write about AI — you wrote with GPT-5.

Walk me through what that collaboration actually looked like. Was there a moment where the AI surprised you, or maybe pushed back in a way you didn’t expect?

Jamie Metzl:

Yeah, it’s a really great question and it’s a really essential question.

So I’ll start at the end and then go backwards.

This book will be the first major trade nonfiction book ever published in human history with a human and an AI as the listed co-authors.

And when people hear that, they imagine there must be some magic AI button.

Like I pressed the button, flew to Florida, hung out on a beach, came back a week later and suddenly there was a book.

Jeff Kavanaugh:

The Jetsons.

Jamie Metzl:

Exactly.

Or honestly, most celebrities use ghost writers.

But what I had was a much deeper, layered, iterative creative process with thousands of back-and-forth interactions with GPT-5.

Jeff Kavanaugh:

And just to put a punctuation point on that — we’re living that at the Knowledge Institute as well.

On an article level, there are dozens of back-and-forth exchanges doing something very similar.

We all have our own assistant now, and if we don’t, that’s a real weakness.

Next, let’s talk about conflicts and similarities.

You drew from religious traditions, indigenous wisdom, even secular philosophy — traditions that often contradict or conflict with each other.

How do you find universal principles without flattening the real differences?

Jamie Metzl:

If some Martian came down to Earth and looked at us, they’d say:

“Wow. It’s incredible. You all have these different traditions, but they all emerge out of human biology and human culture.”

They’re all trying to answer the same questions.

How do we live just lives?

How do we build strong communities?

How do we support our families?

And so for me, there’s so much commonality across our traditions.

But there’s one strict line.

Any tradition that recognizes the validity of other traditions — I think that’s healthy.

But if a tradition says another person must be absolutely wrong for me to be right — especially if it justifies harming people because they think differently — that’s beyond the pale.

Other than that, I think we can recognize each other’s humanity.

None of the AI 10 Commandments are making claims about who’s the son of God or who’s the greatest prophet.

These are principles about how humans can interact with each other and the world around us.

Jeff Kavanaugh:

Business lens here.

Our audience is largely enterprise and technology leaders.

When you look at these commandments as a framework for corporate AI governance — not just societal governance — what becomes directly applicable?

Jamie Metzl:

That’s such an important question.

Every company and every employee needs to think about their education plan.

Not just what they need to know — but what’s adjacent to what they know.

And what I tell people is this:

If you are thinking about how to apply technologies and the first question you ask yourself is, “How do we apply these technologies?” you will have failed before your process began.

Because the first question has to be:

Who am I?

What do I stand for?

What am I trying to achieve?

Jeff Kavanaugh:

So Simon Sinek’s purpose framework.

Jamie Metzl:

Exactly.

Then comes: “How do I do it?”

And once you know who you are and what you stand for, then you decide what are the human parts of the work, what are the machine parts of the work, and what’s the relationship between the two.

That’s why you’re seeing companies like Moderna combine AI and human resources into one department.

Because it’s all about how work gets done.

Machines should do machine things.

Humans should do uniquely human things.

And both should align around shared values and principles.

Anthropic now has an AI constitution.

I think that’s important.

It’s easy for companies and individuals to get lost when integrating powerful technologies like AI.

Starting with principles matters.

Jeff Kavanaugh:

You’ve been vocal about the failure of global AI governance.

The comparison to the early nuclear era comes up often.

Are we closer to Geneva Conventions for AI, or are we falling further behind?

Jamie Metzl:

We’re falling further behind.

These are incredibly powerful technologies.

But effective governance isn’t about slowing things down.

It’s about moving at the right pace with the right safeguards so the whole thing doesn’t blow up in ways that harm us.

Jeff Kavanaugh:

The book is described as deeply hopeful — and that’s a deliberate choice.

At a moment when a lot of serious thinkers are genuinely frightened, what’s the source of your optimism?

Is it earned optimism, or just force of will?

Jamie Metzl:

I need to confess that I’m an innate optimist.

I’ve been smiling since I was a little kid.

And when we look at the arc of history — however uneven — this is still the best time ever to be alive.

If we don’t completely screw things up, tomorrow can be even better.

And now we have these incredible technologies that can help us advance healthcare, energy, agriculture, materials science — all kinds of things.

There are very real risks.

But the reason I write these books is because I want to invest in the positive future without being Pollyannish about the dangers.

The downsides are real.

So we need to invest in the good and fight against the bad.

Everybody has a role in tipping the balance.

Jeff Kavanaugh:

Speaking of Superconvergence — another excellent book — you argued that AI, genetics, and biotech aren’t separate revolutions.

They’re one.

How does the moral framework in this book apply across that convergence?

And are 10 commandments enough?

Jamie Metzl:

When you go into nature, you don’t say:

“Where’s the physics? Where’s the chemistry? Where’s the biology?”

There’s just nature.

These categories are simply our way of organizing understanding.

And it’s the same with technology.

All technologies are embedded in all other technologies.

We’re coming together globally and civilizationally.

And we need some shared sense of common humanity.

Different traditions have tried to build those frameworks.

But there’s something powerful about using a relatively neutral technology to examine the collective wisdom of humanity and ask:

“What principles unite us?”

Jeff Kavanaugh:

As we move toward a closing and call to action:

If a business leader or policymaker reads this book and wants to do something tomorrow — on Monday morning — what’s one concrete thing they should do differently?

Jamie Metzl:

There’s so much wisdom already inside organizations.

Not just wisdom about values — wisdom about everything.

So what I recommend is this:

Read the principles.

Share them with your teams.

And start conversations around two questions.

First: What are the animating values that should guide all of our work?

Second: When we think about the transformation coming from AI and interconnected technologies, how should we approach it?

That’s why I think models like Shopify are really smart.

They started with shared principles.

Then they asked their teams to identify specific problems AI could solve.

And if AI freed up human time, what uniquely human work could people do with that new capacity?

That’s the real goal.

Not humans giving up agency.

Humans should offload to machines the things machines do better.

And then over-index on the things humans do better.

A lot of people talk about AGI — artificial general intelligence.

But I summarize my views on AGI in seven letters:

“AGI is BS.”

If people mean AI systems that can do many things, we already have that.

But if they mean systems that can fully replace human intelligence, I don’t think we’ll ever have that.

Human intelligence is embodied.

It’s the result of nearly four billion years of evolution.

I’m a humanist.

And I think the most successful companies will be the ones that recognize the value of machines doing machine things — and humans doing deeply human things.

Jeff Kavanaugh:

Jamie, this is exactly the kind of conversation our listeners come here for.

Big ideas, grounded in reality, and genuinely useful.

Jamie, thanks so much for being on the Knowledge Institute podcast.

Jamie Metzl:

My great pleasure, Jeff.

Jeff Kavanaugh:

The AI 10 Commandments: A New Moral Code for Humanity is out now.

You can find details at Infosys.com/IKI in our podcast section.

And everyone, be sure to follow us wherever you get your podcasts.

Yulia De Bari and Christine Calhoun produced this podcast.

Dode Bigley is our audio engineer.

I’m Jeff Kavanaugh, and from the Infosys Knowledge Institute — until next time, keep learning and keep sharing.