Labor Day began as a public recognition of the people whose work built the country's strength, prosperity, and well-being. In 2026, that recognition arrives during another reordering of work.
Artificial intelligence can now research, write, code, analyze, design, operate software, and carry increasingly complex assignments. Automation is moving from the factory floor into every department. Owners are measuring tasks that once seemed inseparable from a job title. Workers are wondering whether the next system will make them more capable or simply make them less necessary.
Those are not abstract concerns. They are boardroom decisions, family decisions, and increasingly national decisions. But Labor Day gives us a useful way to frame them.
The purpose of technology should not be to remove people from the story. It should be to remove friction, waste, danger, and repetition so that people can contribute where judgment, courage, trust, and imagination matter most.
A holiday created to recognize contribution
The first Labor Day celebration took place in New York City on September 5, 1882. The U.S. Department of Labor traces the holiday to labor activists who wanted a public celebration of workers and their contribution to American life. Oregon became the first state to formally recognize the holiday in 1887. Congress made the first Monday in September a legal holiday in 1894.
The origin matters because Labor Day was not created to celebrate activity for its own sake. It recognized what work produced: communities, infrastructure, mobility, prosperity, and a better standard of living. It also recognized that progress came with obligations to the people creating it.
That distinction is especially important now. We often measure work through hours, headcount, tickets closed, campaigns launched, or documents produced. AI exposes how incomplete those measurements are. If a machine can produce one hundred drafts, the value is not in reaching one hundred. The value is knowing what should be said, what is true, what serves the customer, what creates risk, and what deserves to exist at all.
AI is changing tasks faster than it is changing the meaning of work
The labor conversation often collapses into two extremes. One side treats AI as a universal job destroyer. The other treats every concern as fear of progress. Reality is more complicated.
The International Labour Organization's refined global index estimates that one in four workers is in an occupation with some exposure to generative AI. Clerical work remains the most exposed, while increasingly capable systems are expanding exposure across professional and technical fields. Exposure, however, is not the same thing as immediate replacement. A job is a collection of tasks, relationships, responsibilities, and context. Some parts can be automated long before the whole role can be responsibly redesigned.
The Bureau of Labor Statistics now maintains dedicated research on AI and productivity. That is exactly where business leaders should focus. The strategic question is not whether a model can imitate a task during a demonstration. It is whether a redesigned system can produce a better, safer, more economical outcome over time.
This is the same transition I explored in Fable 5.1 and GPT-6 Astra Are Here. Frontier systems can carry more of an assignment, but capability alone does not create a responsible operating model. The business still needs clear objectives, authoritative data, permissions, evaluation, human ownership, and recovery when the system is wrong.
Productivity is valuable only when its benefits become useful
The OECD's 2026 productivity indicators report that labor productivity grew across much of the OECD and note that AI may be contributing to gains despite economic uncertainty and skill shortages. Productivity growth matters. It can lower costs, expand capacity, improve living standards, and free resources for new investment.
But productivity is not automatically humane, broadly shared, or even durable. A company can automate a poor process and produce mistakes faster. It can use monitoring technology to increase pressure while reducing autonomy. It can remove experienced people before capturing the judgment that made the process work. It can claim savings while pushing hidden costs onto customers, remaining employees, or the future.
OECD research on AI, performance, and job quality points to the better path. AI use can be associated with stronger engagement, learning, flexibility, and improved performance, particularly when companies use it to support and empower workers. The same technology can produce worse outcomes when it raises work intensity and reduces human autonomy.
THE RESPONSIBLE AUTOMATION TEST
- VALUEDoes the outcome improve?Measure quality, safety, speed, cost, customer experience, and resilience together.
- PEOPLEDoes human capability grow?Move people toward judgment, relationships, craft, creativity, and accountable decisions.
- CONTROLIs authority bounded?Give systems only the data, tools, permissions, and time required for the assignment.
- SHAREWho receives the benefit?Turn productivity into stronger service, better work, investment, opportunity, and durable value.
Good automation puts power behind human capital
The strongest AI systems do not begin with the instruction to replace a department. They begin by mapping the work.
Where does information stall? Which decisions depend on unavailable context? What repetitive activity prevents experienced people from serving customers? Which processes exist only because software has never understood natural language, images, intent, or exceptions? Where is risk concentrated in one exhausted person who has become the human integration layer between five systems?
Once the work is visible, leaders can make deliberate choices. Machines should carry volume, repetition, retrieval, monitoring, first-pass synthesis, and tightly bounded execution. People should own purpose, exceptions, ethical judgment, negotiation, trust, invention, and responsibility.
This is why I argue in The Cognitive Advantage that the real advantage comes from faster sensing, better interpretation, and controlled action. AI is powerful because it can help information move. Human capital becomes more valuable when people receive the context and time required to make the decisions only they should make.
The next labor advantage belongs to learners
Every major technology transition changes the premium placed on particular skills. The web reduced the cost of publishing and distribution. Cloud computing reduced the cost of infrastructure. Smartphones reorganized access, commerce, media, and attention. AI is reducing the cost of applying certain forms of intelligence to a task.
That will pressure roles built primarily around transferring information from one format to another. It will also create extraordinary leverage for people who understand customers, systems, risk, operations, and the practical limits of the tools.
The useful response is neither panic nor passive optimism. Learn how the technology works. Use it on real assignments. Understand context, tools, memory, permissions, evaluation, and cost. Become the person who can define the right outcome and recognize when the machine has produced the wrong one.
That learning cannot be reserved for executives and engineers. Organizations should give their people safe environments, representative problems, clear policies, and time to build fluency. A company that automates around its workforce without teaching that workforce will create resistance and lose insight. A company that invites its best people to redesign the work can capture both operational knowledge and new capability.
Leadership must decide what progress is for
Technology does not decide whether productivity becomes layoffs, lower prices, better service, shorter workweeks, stronger margins, new products, or reinvestment. Leaders do.
There will be legitimate reductions where an old operating model no longer makes sense. Pretending otherwise is dishonest. There will also be companies that cut too deeply, surrender institutional knowledge, and discover that an automated process without experienced judgment is simply a faster route to failure.
The leaders worth following will communicate clearly about what is changing and why. They will measure the human consequences alongside the financial ones. They will preserve accountability even when a machine performs the action. They will recognize that efficiency is a means, not a mission.
As I wrote in Digital Marketing Agencies Are Dead, automation can eliminate enormous waste in fragmented operations. The goal is not to make people compete with software at repetitive work. It is to build connected systems that let strong people apply their knowledge with greater reach.
PUT POWER BEHIND HUMAN CAPITAL
- 01Map the real workDocument decisions, handoffs, exceptions, delays, risks, and the knowledge people carry.
- 02Remove the frictionAutomate repetition, retrieval, reconciliation, monitoring, and predictable execution.
- 03Elevate the personShift time toward customers, judgment, invention, craft, leadership, and trust.
- 04Govern the machineRequire evidence, permissions, review, auditability, and a clear owner for every outcome.
- 05Share the gainConvert productivity into opportunity, resilience, better service, and meaningful growth.
A Labor Day wish for the age of intelligence
Today, I am grateful for the people who build, repair, teach, protect, care, design, code, lead, deliver, serve, and solve. I am grateful for the tradespeople whose work is visible at the end of every day and for the knowledge workers whose best contribution may be a decision that prevents a problem anyone else ever sees.
I am also optimistic about what good technology can do. It can make dangerous work safer. It can make expertise available sooner. It can help small teams compete with large institutions. It can give a talented person the operating leverage that once required an entire department.
But that future is not guaranteed by the model. We have to design it.
Happy Labor Day. May the work be useful, the progress be shared, and the people behind it remain visible.
Continue through the connected system: Read how AI changes organizational advantage, examine what frontier models mean for the workforce, and learn more about Brad's career building systems across technology, business, and national security.
Sources and further reading
- U.S. Department of Labor: History of Labor Day
- U.S. Bureau of Labor Statistics: Productivity and Artificial Intelligence
- International Labour Organization: Generative AI and Jobs
- OECD: Compendium of Productivity Indicators 2026
- OECD: Reaping the Benefits of AI for Performance and Job Quality
Put intelligence behind your people.
If you are trying to reduce operational waste, redesign work, or introduce AI without sacrificing quality and accountability, start a conversation with Brad. We can map the operation, architect the intelligence layer, and build governed automation that makes human capital more capable.