OpenAI Wants to Reshape Industrial Policy. Warehouse Workers Should Be Paying Attention.

The company didn't provide specifics on workforce transition programs or labor protections. That's not unusual for these kinds of announcements. Tech companies tend to speak in generalities when it comes to the human cost of automation. They talk about "new opportunities" without explaining what happens to the people whose current opportunities disappear.

What "Dramatically More Capable" Actually Means

Altman's framing is interesting. He argues that AI will make people dramatically more capable, that the biggest problems of today will become solvable. In a companion piece, OpenAI introduced what they call "AI stories," meant to highlight daily benefits that point toward bigger opportunities.

I don't want to be cynical about this. Some of it is probably true. AI-assisted diagnostics might catch cancers earlier. Language models might help students learn in ways that weren't possible before. These are real benefits.

But here's the thing: the workers I've talked to over the years aren't worried about whether AI will solve abstract problems in science and medicine. They're worried about whether they'll have a job in two years. Whether the skills they've developed, the physical knowledge of how to move efficiently through a 500,000 square foot facility, will matter when the robots get better.

And the robots are getting better. That's not speculation. You can see it in the deployment numbers, in the press releases from automation companies, in the changing layouts of fulfillment centers where human-accessible aisles are being replaced by dense robotic storage systems.

The Prosperity Question

OpenAI talks about "sharing prosperity." This is good for the company, but what about the people? It's a question that doesn't get asked enough in these conversations.

The warehouse industry employs roughly 1.9 million people in the United States, according to Bureau of Labor Statistics data. That's a lot of families, a lot of communities that depend on these jobs. Many of these positions don't require a four-year degree, which makes them accessible in ways that other jobs aren't. They're entry points into the middle class, or at least they used to be.

When companies talk about the Intelligence Age, they're usually talking about knowledge workers. Programmers, researchers, analysts. People who sit at desks and use AI tools to augment their thinking. That's a real thing, and it matters. But it's not the whole picture.

The whole picture includes the night shift at a fulfillment center in Joliet. It includes the workers who load trucks at 4 AM so that packages arrive on time. It includes the people whose bodies bear the cost of same-day delivery.

I'm not saying OpenAI has to solve all of these problems. They're an AI company, not a labor policy organization. But when you publish proposals for reshaping industrial policy, you're making claims about what kind of future you want to build. And if that future doesn't have a clear place for the people who currently make the economy run, then the rest of us should be, well, skeptical.

What's Missing

Here's what I didn't find in OpenAI's proposals: specific commitments to worker retraining programs. Partnerships with unions or labor organizations. Acknowledgment that the transition to an "Intelligence Age" might be painful for millions of people, and that pain isn't equally distributed.

Maybe those details are coming. Maybe this is just the opening statement in a longer conversation. I hope so.

But right now, we have a company valued at over $150 billion talking about expanding opportunity and sharing prosperity without explaining how that sharing is supposed to work. Without acknowledging that, historically, the benefits of technological change have gone mostly to the people who already had capital and education, while the costs have been borne by the people who didn't.

It's too early to say how any of this will play out. The AI systems that OpenAI and others are building might create entirely new categories of work that we can't imagine yet. That's happened before. The automobile eliminated jobs for blacksmiths and stable hands but created jobs for mechanics and assembly line workers.

Or it might not happen that way. The transition might be faster and more disruptive than previous technological shifts. We don't know yet.

Watching and Waiting

The workers I spoke to last year at a facility outside of Memphis had a phrase they used: "watching and waiting." They knew automation was coming. They could see it in the new equipment being installed, in the training sessions for the supervisors, in the way management talked about "optimization" and "efficiency."

They weren't opposed to technology. Most of them used smartphones, ordered things online, understood that the world was changing. But they wanted to know what the plan was for them. Not in abstract terms, not in policy papers about the Intelligence Age, but in concrete terms. What happens to me? What happens to my coworkers? What happens to this community if this facility closes or reduces headcount by 40 percent?

Those are the questions that OpenAI's proposals don't answer. And until they do, all the talk about expanding opportunity and sharing prosperity is just that. Talk.

I'll be watching to see what comes next. Whether this is the start of a genuine conversation about how to manage the transition to a more automated economy, or whether it's just another tech company promising a better future while the people who build and move things are left to figure it out on their own.

The view from the floor is different from the view from San Francisco. Someone should probably mention that.

James Chen · 25 Jun · 5 min