It’s time to start planning for AI job losses

I’m frustrated.

If you weren’t already aware of the AI-powered calamity in our near future, Matt Shumer gave it a good intro in “Something Big is Coming.”

We’ve been warned, but we’re not taking the warning seriously.

AI is already causing mass layoffs, and reports estimate anywhere from 10 to 50% of jobs could be automated in the next decade. I think that’s on the low end.

Intelligent people keep telling me that AI will work out just fine, but all they can give me is vague, hand-waving, happy clappy optimism. There’s meat in the negativity and whipped, sugar-free foam in the positive spin.

“AI will create so much prosperity that everyone will be wealthy.”

That’s a lovely sentiment, but have you thought it through? You might be able to argue that if AI drastically lowers the cost of goods then people’s money will go further, which does indeed create wealth. Unfortunately, that assumes everything else remains the same. For example, that people still have a source of income.

The truck driver who’s displaced by a self-driving truck isn’t impressed that the cost of goods is falling. He doesn’t have the money to pay for those goods even at the reduced price.

The AI tsunami is on its way, and it’s going to displace a lot more than truck drivers. Customer service reps, legal assistants, junior programmers, accountants …. Lots of workers are looking into the abyss. We don’t have time for the sloppy thinking that has prevailed.

If you’re going to sell a theory about why AI won’t destroy civilization, please have the courtesy to explain the mechanism that makes it plausible, and please tell us that someone is working on putting that mechanism in place.

For example …

If you believe AI will raise everyone’s standard of living, you must know how people with no jobs are going to buy things.

Maybe you think Universal Basic Income will help, so …

If you believe the increased wealth from AI will be taxed and redistributed, you must know how governments will identify which earnings are from AI productivity gains and how they’re going to tax them.

You must also know how AI can be simultaneously creating all this wealth and paying all these taxes.

If you believe AI will create more jobs than it destroys, you must have some idea what those jobs might be, and how they compare in scale to the jobs that are lost.

If you believe humans will always outperform machines in “the important things,” you must believe “the important things” will outnumber the stuff AI can do better.

If you believe out-of-work humans will be freed to be creative or study philosophy, you must be talking about some other variety of humans. A large portion of the population will watch TikTok videos, get fat, and abuse drugs.

The point is simple. Optimism is not an argument. Hope is not a strategy or a mechanism.

Listen. I’m a Star Trek fan. I would love to live in a world where you get your food from a replicator and you only work (or cook) if you want to. If that’s the world AI is going to create, I’m in. The question is how we get from here to there without a hideous calamity along the way.

If you want to have a rational hope for a future where AI helps humans to prosper, you need to think past the naive talking points and focus on practical mechanisms.

Or were you hoping the politicians would figure that out? (Seriously, were you thinking that?)

We can’t pin our hopes on politicians, tech titans, geniuses, or experts. That’s not how actual progress works.

Have you heard of Gall’s law? It says that a complicated system can’t be designed from scratch. It has to be designed from simpler systems that have proven themselves to work in the real world.

To get through the transition that’s flying towards us at warp speed, thousands of theories have to be tested in the real world so that the few that work can rise to the top.

Think back to the pandemic. The happy talk theory was that with all that time on their hands, people would learn an instrument, study philosophy, or create works of art.

Yeah. Five people did that. The rest were drinking gin on Zoom calls, pretending it was water.

AI is here, and it’s only going to get better and more capable. We can hide our heads under a blanket and wait for the end, we can lead some sort of hopeless jihad against the thinking machines, or …. We can try to find a way to make this work.

I have grandchildren. I want them to live in a better world, so there’s only one real choice for me.

I’ve been torturing myself with this question for months. How can humans prosper in an age of AI? What methods, mechanisms, and mental models might work, and how can we start testing them right now?

I’m a creative guy, and I’ll do my part, but I can’t come up with all the ideas we’ll need. The crisis we’re facing requires a million creative people who (1) realize the horrible danger we’re in and (2) are willing to look it in the eye and stare it down.

We need to start imagining how humanity is going to work with AI. Practically. In a way that includes things like paying bills and buying groceries.

Here are some seed ideas to get the conversation started.

What if CEOs started thinking this way?

• Make a commitment that all AI productivity gains are reinvested in new initiatives, thus fulfilling the promise that AI creates more jobs than it destroys.
• Give every employee one hour a day to come up with new ideas. (BTW, you’ll need a process to sort those ideas. I’m good at that.)
• Allow / encourage employees to work side jobs to diversify their personal revenue streams.
• Experiment with shorter hours instead of fewer people – while maintaining salaries.
• If AI allows the same work to be done with fewer people, use the surplus time for training or mentoring.
• Change the corporate culture so that CEOs take lower compensation to keep people on the payroll.

But who am I kidding? Corporations are driven by greed. We can’t rely on the CEOs any more than we can rely on the tech bros. Or Congress.

We have to start thinking about changes on the personal level. Here are some ideas.

• Do business with the (few) companies that implement human-centric policies.
• Live well below your means so you can share with others who are victims of the tsunami.
• Insist on high-quality products with a human touch. Don’t settle for AI slop.
• If losing a job would lead to a loss of meaning or purpose, start looking for other ways to find fulfillment in your life.
• Invest in your community. If we’re all out of work, you’re going to be hoeing turnips next to your neighbor. You might as well get to know him now.
• Diversify your income streams. If you don’t have a side gig, start one.
• When you think of a good idea for a product or service, use AI to make it.

None of this requires permission. None of it depends on a tech breakthrough or a new government program. These are ideas ordinary people and companies can try. And I haven’t even mentioned education, culture, art, etc.

Be clear-eyed about the challenge. Most ideas fail. But some will work better than expected for reasons nobody could anticipate. Cast your bread on the waters. Try things, and encourage others to do the same.

The need of the moment is both stark and clear.

• A realistic assessment of the storm that’s coming.
• Resolve to face it.
• Creativity to beat it.
• Sacrifice to get through the hard times.

Are you in, or are you under your blanket?