I Thought My Working Life Was Over
I was 46 when the plant manager gathered us together and told us the company was installing a new generation of automated manufacturing systems. He said it was necessary to remain competitive. He said there would be new opportunities. What I heard was: They don't need you anymore. I'd worked there for 23 years. I wasn't poor. I wasn't rich. My wife, Denise, worked three days a week at a dental office. We owned a modest house with a mortgage. We had two teenagers, one getting close to college. I was good at my job. Then suddenly being good at my job didn't seem to matter.
I was angry before I was scared
At first I blamed AI. Then management. Then politicians. Then younger workers. Mostly I was terrified. I'd been running manufacturing equipment since I was 23. What exactly was I supposed to become at 46? Denise suggested HOS. I told her I didn't need a computer telling me how to live my life. Three weeks later I signed up. My first question was: “What job can I get that pays what I'm making now?” HOS didn't have a good answer. I didn't like that. Instead, we began figuring out what was actually happening.
How quickly would my job change? Which tasks were disappearing? Which weren't? What did the new equipment require? What did I know after 23 years that wasn't written in an operating manual? That last question changed things. I knew what a machine sounded like just before something went wrong. I knew which production problems were really equipment problems and which were people problems. I knew how one small change upstream could create a mess three stations later. The new AI knew things I didn't. But I knew things too.
I made my first mistake
HOS helped me identify a company training program for technicians working with the new automated systems. I enrolled. I lasted four weeks. Everyone seemed younger. The terminology was foreign. The computer work embarrassed me. So I quit. I told Denise the program was stupid. She didn't argue. She just asked: “Was it stupid, or did it make you feel stupid?” I was furious because I knew the answer. That night I went back to HOS. I expected encouragement. Instead, we examined what had happened. I hadn't failed at learning automation. I had tried to learn too much, too quickly, in an environment where I was ashamed to admit what I didn't understand.
Those were different problems. So we built a different approach. I learned some basics privately. A younger technician at work showed me things. I showed him things about the production line that he didn't know. Six months later I enrolled again. I finished. Failure hadn't proved I couldn't do it. It had taught me something about how I needed to learn.
Then I almost made a bigger mistake
Two years later I was offered a supervisory job. More money. Better title. I was ready to say yes. HOS helped me look beyond the raise. The job meant nights. Denise and I were already having trouble finding time together. Our daughter was struggling. Our son was applying to college. I took the job anyway. I wanted to prove something. Within a year I was miserable. For months I blamed everybody else. Eventually I admitted that the promotion had been my decision. HOS had helped me see the tradeoffs. I had simply chosen differently. That became an important lesson. Having better information doesn't guarantee a better decision. Agency includes the freedom to make mistakes.
It also includes the ability to learn from them. I eventually moved into a technical training role. It paid slightly less than the supervisor position and fit me much better. I discovered I loved teaching people who were afraid they couldn't learn. I knew that feeling.
The hardest automation wasn't at the plant
By my mid-fifties, manufacturing looked completely different. There were fewer people on the floor. The people who remained did different work. Some of my old friends never recovered from the transition. One bounced between jobs. Another developed a drinking problem. One retrained faster than I did and ended up earning more than any of us. HOS couldn't guarantee outcomes. Neither could I. But it helped me keep asking: What has changed? What hasn't? What do I understand? What do I need to learn? What choices do I actually have? What happens if I do nothing? That last question mattered. Sometimes doing nothing was also a choice.
Then my son lost his job
Years later, my son called me. AI had eliminated much of the work his company had hired him to do. He was devastated. I started giving him advice. He got irritated. I couldn't understand why. Then I remembered myself at 46. I was trying to solve his life because I loved him. But it was his life. HOS helped me see the difference between supporting someone's agency and taking over their decisions. So I asked him: “What do you think you want to do?” He said: “I don't know.” I smiled. “That,” I told him, “is actually a pretty good place to start.”
Looking back
I retired at 68. Not from the job I expected. Not with the career I planned. That career disappeared years earlier. But my working life didn't disappear with it. Something strange happened instead. The technology that frightened me forced me to learn how to learn again. That ability became more valuable than the technical skill I originally went back to school to acquire. I used HOS through four major changes in my work. It never predicted them all. It never guaranteed that I would succeed. And sometimes I ignored what I learned and made the wrong choice anyway. But gradually I stopped believing that losing an old capability meant losing my usefulness.
The world would change again. So would I. At 46 I thought resilience meant surviving what happened to me. By the end of my working life, I understood it differently. Resilience was becoming capable of creating another choice when the choice I expected was no longer available. AI changed my job. It did not get to decide my life. I did. And I'm glad I did.
Here HOS begins to show another dimension of agency: resilience. When technology makes an established career less valuable, the objective is not simply to identify the next job. HOS helps the worker understand what is changing, recognize capabilities he may not have valued, discover what he needs to learn and take action. Importantly, the story allows him to fail. He quits his first training program, later accepts a promotion that proves wrong for him, and sometimes ignores what he already understands. HOS does not protect him from those choices or manufacture successful outcomes. It helps turn experience—including failure—into deeper understanding and stronger future judgment. As AI transforms work, another AI system might become extraordinarily capable of predicting which occupation, training program or promotion is statistically optimal. HOS can use such intelligence, but its deeper purpose is to strengthen the human being's capacity to keep learning and adapting when predictions fail or circumstances change. Resilience here is not simply recovering from disruption. It is becoming increasingly capable of creating another meaningful choice when the choice you expected is no longer available.