One Thing Humans Do Better Than AI

…is die.

Stay with me. I know that’s a strange way to start a pep talk. 

The more conventional pep talk to humans these days centers on our hard-to-replicate judgement—the kind of discernment that comes from years of roving the earth in a fleshy body. The technology is evolving fast though, and there are compelling arguments out there that (ack!) even judgement may not be uniquely human anymore.

But while Silicon Valley has thrown everything it has at immortality for decades, humans are still really great at dying. As far as I know, we continue to have a ::checks notes:: 100% success rate on this front. 

Why is our pesky mortality an enormous advantage as we figure out* how to co-exist with a new digital species**? Because we’re the ones who own meaning.

We’re the ones painfully aware that our time on this planet is finite—and most of us want that time to mean something. Not only do we want to nurture those closest to us; we want to see the world get better. We want abundance not scarcity—in water, food, shelter, health, safety, learning, innovation, opportunity, etc. 

In my view, this will only happen if we mere mortals keep the upper hand—if we (not just humans, but the vast swath of ordinary humans) remain the deciders. 

There are so many critical choices ahead of us. For example:

• How will we make whole the millions of people whose original work was used to train LLMs—without consent, credit, or compensation? 
• How will we continue to incentivize original human work going forward?
• What social safety nets will we introduce as job dislocation inevitably ratchets up?
• How should we change our tax structure—lest an unprecedented concentration of wealth takes hold before we can undo it? 
• How will we re-think education—both for kids and for adults?
• How will we keep people sane in a world so deeply disorienting?

I get it: not everyone can think about these concerns from dawn until dusk. Most people are just trying to get through the day. Most people are just trying to make ends meet. As the owner of a mid-sized business, I have a particular soft spot for the leaders of small companies and non-profits. At these organizations, staying afloat is an all-consuming task. For their health, and the health of their organizations, I usually advise such leaders to focus on the most pressing problems at hand—the problems over which they have the most control. 

That said, we can’t let all our leaders and organizations focus solely on the near term. We need leaders who are thinking ahead and looking out for society as a whole. To that end, I stand by these seven questions that I believe every presidential candidate—heck, every politician!—should have to answer. And I am grateful for the work of places like the Windfall Trust, which is grappling with the economic and social implications of AI’s current trajectory. 

I learned about this organization early this morning when I saw that a friend, the Pulitzer Prize-winning journalist Mei Fong, had chosen to join their ranks. For the interests of the greater good to prevail, it’s going to take millions of us pulling in the same direction, but—in the metaphorical gymnasium where sides are being sorted—Team Mortal just got a top draft pick.

Notes:
*I know there is a vocal movement to just “reject AI.” I think this makes no sense for multiple reasons, all of them deeper than the oft-heard and very lazy, “That horse is out of the barn.” For one thing, while generative AI hit the mainstream only in recent years, we have all lived with all manner of artificial intelligence for decades. And much of it is an unequivocal good. Trust me: you do not want me reverting to paper maps while I drive. 

**I know there is an equally vocal movement to reject the concept of AI as a digital “species.” I understand these objections, and I’m open to a better label—but I reject the idea of AI as a tool. The ‘tool’ framing leads us to vastly underestimate the abilities—and risks—developers have unleashed. If we have any hope of putting the necessary guardrails in place, we need our legislators—yep, the very people who asked Mark Zuckerberg to explain the internet to them—to understand that AI has agency in a way that no other tool ever has.