TL;DR: The 71-year-old memoir client I work with uses AI more thoughtfully than most professionals half his age. He has it find missing bits in his memoir draft, refocus him when he loses the thread, and write notes for his ghostwriter about what to work on next. He never lets it write a word of the book, and he’s a better writer now than he was five years ago, partly because of the tool.
He sits down at his desk a little after 7 in the morning, every day, the way he has for sixty years of work. Coffee on the right. A pair of reading glasses on the open notebook in front of him. Two screens, one for the manuscript in progress, one for the AI tool he’s been using for the last eight months.
He’s 71 years old. He’s writing a memoir of a life that filled three careers and two marriages. A country he left and another he made his home in. And a stretch of years he doesn’t talk about easily and is now writing about for the first time.
And he’s using AI better than almost any of my younger clients.
I like his approach more than almost anything I hear from either side of the AI argument. The boosters want the machine to write the book, and the critics want writers to pretend the machine doesn’t exist. AI is a tool. Treat it like one and you’ll do fine. Let it do your work for you and you’ll have problems. He’s never once confused the two.
The work
The memoir is long. We’ve been on it for the better part of a year, and we’ll be on it for a while longer. Some weeks the work is mostly his, drafting the parts he wants to write himself. Some weeks the work is mostly mine, interviewing him about a chapter and then writing it from his voice. We trade the draft back and forth, and the manuscript thickens slowly.
Every word in his manuscript is his, because the book is his life and the machine has no business inventing any of it.Share on X
What changed this year is how he prepares for our sessions. Eight months ago he started running his own draft chapters through AI before our calls. Not to have the AI rewrite anything. To find the parts of the draft that weren’t working.
He asks the AI to read a chapter and tell him where the pacing falls off. Then he has it flag passages where he might be assuming the reader already knows something he hasn’t told them. He asks it to find places where two chapters seem to be telling the same story from two different angles, when one of them probably needs to be cut.
He asks it to identify the moments where his voice changes, where a section reads like a different person wrote it. That usually means he was tired when he wrote that part and should rewrite it. He doesn’t take the AI’s suggestions at face value. He reads them, decides which ones land, and brings the rest to our session as questions for me. Half of what he flags ends up being useful. The other half he discards. The work is his.
I wish more writers understood that throwing out half of what the machine says is the skill. A writer who accepts every suggestion ends up with a book that sounds like everybody else’s, and he can’t hear it himself.
What does a 71-year-old memoirist use AI for?
Three things. I asked him directly the last time we met.
The first is finding missing bits. He says he can read his own draft for the fifth time and not see the gap. The AI doesn’t have his memory of what he meant to say, so when something is missing, the AI feels the gap. It tells him “this section refers to your father, but you haven’t introduced your father yet at this point in the book.” He looks at the section and realizes he wrote two chapters in a different order from the one he intended and the introduction got moved. He fixes it.
The second is focus. Some chapters he’s worked for weeks, until he can no longer tell what the chapter is about. Then he asks the AI to read it and tell him in one sentence what the chapter is doing. The answer is sometimes what he intended. Sometimes it isn’t. When it isn’t, he knows the chapter has drifted from his intent, and he goes back and rewrites it toward what he wanted to do.
The third is generating notes for me. Before our sessions, he runs the latest version of a chapter through the AI and asks it to summarize what’s working, what’s weak, and what he wants to discuss. He sends me that summary. I read it before the call. The hour we spend together is now a focused conversation about specific moments instead of a wandering attempt to find what we should talk about. The work goes faster, and the questions he asks me are sharper.
I love getting those notes. A client who shows up knowing what he wants to talk about gets far more out of an hour with me than one who spends forty minutes hunting for the question, and the book gets better for it.
What doesn’t the 71-year-old memoirist use AI for?
He’s never let the AI write a single sentence of his book. He’s adamant about that. I asked him why. He said the book is the only place where the voice is fully his. The moment he lets the machine write a sentence, the voice starts being something else. He said it like it was obvious. Seventy-one-year-olds get to do that.
The line he draws is the same line every working writer I respect draws. The AI handles the logistics. The human writes the prose. Every word in his manuscript is his, because the book is his life and the machine has no business inventing any of it. The full version of why this matters is in It’s Not the AI, It’s the Crap, but the shorter version is the one he gave me.
Why this works for him
A few reasons. He’s not afraid of the technology because he’s been around long enough to watch a dozen technologies arrive and leave or stick. He didn’t start the memoir to make money from it. He’s not trying to compete with anyone, so he isn’t tempted to take a shortcut that would hurt the work.
I think that last point explains him better than his age does. The people who misuse AI worst are the ones in a hurry to cash in, and a shortcut that saves a week and costs the reader’s trust is a bad trade in any genre.
And he understands the difference between using a tool and being used by one. He hands AI the parts of writing that drain him without giving him anything in return: the looking-for-gaps work, the keeping-track-of-what-he-meant work, the boring logistics of organizing his thoughts so the actual writing can happen. The time he gets back goes into what he came to do, writing the parts of his life he’s been thinking about for sixty years and is finally ready to say.
This is exactly the model from Augmented Beats Replaced. The machine handles the routine. The human handles the work that matters. The work that matters here is a memoir that will eventually sit on his children’s shelves, and the writing has to be his because that’s the point of the book.
What he’s not afraid of
He’s not afraid of AI replacing him as a writer. When it comes up, he laughs about it. He says the AI doesn’t have a life to write about, and even if it learned to mimic his voice, it would be mimicking his voice without having lived the life that produced the voice. He isn’t worried about that, he says.
Too old to learn the tool? He’s not afraid of that. He learned it the way he learns anything, which is by sitting down with it for a few hours, breaking things, and figuring out how it works. The first month he used it badly. The second month he started using it well. By the third month he had it integrated into his workflow. He’s never asked me to explain anything about it to him because he assumed he could figure it out. That’s the right assumption.
Judgment for using AI doesn’t scare him. He doesn’t hide it. When his friends ask whether he’s writing the book himself, he says yes, and then tells them about the AI and how he uses it. They understand the distinction or they don’t. He doesn’t lose sleep over the ones who don’t.
I wish more people half his age had his nerve about this. Plenty of professionals hide their AI use as if it were cheating, and plenty more brag that they never touch it. He does neither, and I think his position is the healthiest one anybody can take.
What he is afraid of
One thing, and it’s the same thing every honest writer of any age has always been afraid of. That he’ll run out of time before the book is finished.
That fear is the one I take most seriously with clients his age, because it’s a real one. A memoir that stalls halfway helps nobody, and a man in his seventies shouldn’t lose a single Saturday to bookkeeping a machine can do for him.
The AI is one of the things helping him not run out of time. It handles the boring overhead that used to consume his Saturdays. The freed Saturdays now go into the writing. The writing accumulates faster than it would have without the tool. The book is going to make it.
That’s the augmented human, at 71, doing the work of his life. Faster, sharper, and more in control of his time than he was before he picked up the tool. The longer version of what this looks like across professions is in The Birth of the Augmented Human, but the version that matters here is the one above, one specific human, in his seventies, finishing a book he wouldn’t have finished without the tool he wasn’t supposed to be young enough to learn.
He’s the model. If you’ve been telling yourself you’re too old, or too busy, or too far behind, he’s the answer to that. Retrain or Be Replaced. There’s No Third Option. walks through the broader argument, but his story is the proof.
People write off anyone his age too quickly as someone who can’t learn this stuff. He taught himself in three months without asking me a single question, and he uses it more carefully than most of the people doing the writing off. Anybody who still thinks age is the obstacle is making an excuse for themselves.
Frequently Asked Questions
The book on this: The Birth of the Augmented Human is 212 pages on what goes right when somebody works with the tools deliberately. That’s exactly what he’s doing at seventy-one.
