I recently read Anthropic’s Economic Policy Framework, a thoughtful attempt to grapple with one of the most important questions surrounding artificial intelligence: what happens if AI begins to replace large amounts of human labor?

To Anthropic’s credit, the paper takes the possibility seriously. That alone is notable.

For years, discussions about AI have largely oscillated between two camps. One camp insists that AI will create new jobs just as previous technological revolutions did. The other predicts widespread unemployment and economic upheaval. Anthropic appears less interested in choosing a side and more interested in preparing for multiple possible futures.

The framework outlines a range of responses depending on the severity of labor market disruption. In milder scenarios, the focus is on retraining workers, helping them transition into new careers, and ensuring that people share in the economic gains created by AI. In more severe scenarios, the conversation shifts toward expanded unemployment support, wealth redistribution, and even concepts such as universal basic income.

Reading the paper, I found myself agreeing with much of it.

And yet I kept returning to a question that sits slightly outside the paper’s central focus.

What if the biggest challenge isn’t economic?

What if it is psychological?

The framework is primarily concerned with income, employment, and economic stability. That makes sense. Policymakers tend to think in terms of measurable outcomes. Unemployment rates can be tracked. Wage growth can be measured. Tax revenues can be projected.

Purpose is harder to quantify.

Meaning rarely appears in a spreadsheet.

Yet when I think about the role work plays in people’s lives, income feels like only part of the story.

Work provides structure.

It creates social connections.

It offers opportunities for mastery and achievement.

For many people, it becomes part of their identity.

When someone introduces themselves as a teacher, nurse, engineer, or electrician, they are rarely describing a source of income alone. They are communicating something about who they are and how they contribute to society.

This is where I wonder whether the conversation around AI remains incomplete.

Suppose Anthropic’s most optimistic scenario comes true. AI dramatically increases productivity. Economic output grows. Wealth expands. Basic needs are met through some combination of wages, public support, ownership structures, or redistribution mechanisms.

Would that automatically produce a flourishing society?

I’m not convinced.

In some ways, we already have a small-scale version of this experiment.

Technology has made many aspects of life easier than at any point in human history. We have access to unlimited entertainment, endless information, instant communication, and increasingly personalised digital experiences.

Yet anxiety, loneliness, and social isolation remain persistent concerns across much of the developed world.

That observation does not prove that technology causes these problems. Human wellbeing is complicated. But it does make me cautious about assuming that greater convenience automatically leads to greater fulfilment.

AI could accelerate this trend.

We are heading towards a world where AI becomes not only our coworker but also our tutor, assistant, companion, entertainer, and advisor.

On one hand, for most people that sounds extraordinary.

On the other, I wonder what happens when more and more human needs are met by systems rather than people.

What happens to community when fewer interactions require other humans?

What happens to ambition when machines can perform many tasks better than we can?

What happens to our sense of usefulness when contribution becomes optional?

A small example of this struck me this morning.

My wife and I were making a packed lunch for our nine-year-old daughter before school. We’d had a busy weekend and were running low on food, particularly the snacks she would normally take with her.

When she found out what was in her lunch, the reaction was dramatic.

At first, it seemed completely disproportionate. After all, she still had lunch. She wasn’t going hungry.

Then it dawned on me that the issue wasn’t really the food.

Packed lunches are part of the social fabric of the playground. Children compare what they have, swap snacks, discuss favourites, and participate in small rituals that help build friendships and a sense of belonging. From an adult perspective, the nutritional requirements had been met. From a child’s perspective, something much more important felt missing.

It made me wonder whether we sometimes make the same mistake when discussing the future of work.

We focus on the practical function while overlooking the social and psychological one.

Work, like that packed lunch, serves purposes that are easy to miss when viewed purely through an economic lens.

The danger may not be that people have nothing to do.

The danger may be that they have endless things to consume and fewer reasons to create.

Historically, human beings have often found meaning through responsibility. Raising children. Building businesses. Serving communities. Solving difficult problems. Caring for others.

Those activities are not always enjoyable in the moment. In fact, they are often difficult.

Yet difficulty itself seems to be part of what makes them meaningful.

If AI removes friction from more areas of life, it may also remove some of the opportunities through which people develop resilience, competence, and purpose.

Of course, there is another possibility.

Perhaps AI frees people to pursue more creative, relational, and human-centred lives. Perhaps the reduction of repetitive work creates space for art, learning, caregiving, entrepreneurship, and community building. Perhaps future generations will look back at our attachment to traditional employment in much the same way we look back at earlier assumptions about work.

That future is entirely possible.

But if it arrives, I suspect it will require more than economic policy.

It will require cultural innovation.

Anthropic’s framework does an admirable job addressing how society might distribute the gains of AI.

The question I kept returning to while reading it is whether we are spending enough time thinking about how society will distribute meaning.

Because if AI succeeds beyond our expectations, the defining challenge may not be how people earn a living.

It may be how people find a reason to get out of bed in the morning.