Most leaders are pretty good at adding things. There’s always room for another initiative, another “important” meeting, or some new process or tech that’s supposed to make everyone’s life easier.
Before long, the organization is carrying far more t…
In 2004, Jeff Bezos banned PowerPoint from Amazon’s senior leadership meetings in favor of narrative memos, read in silence at the start of each meeting. The idea was that a well-structured and tightly argued memo communicated its ideas in a way that a deck built around bullet points could not. As Amazon moves into the AI era, it is developing internal tools that can help produce polished documents much more efficiently. Its internal AI assistant, Cedric, promises to deliver “six-page narratives in seconds,” according to Kristi Coulter, a former Amazon executive.
But no matter how good Cedric’s outputs are, using an AI tool to trim the traditional writing process down to mere seconds undercuts what made well-written memos valuable in the first place. The memo itself was never the point. What mattered was what writing the memo forced the human to do. As Bezos put it, there is “no way to write a six-page, narratively structured memo and not have clear thinking.” The process, not the output, was the real source of value.
How AI breaks the memo
Bezos wasn’t promoting writing for its own sake. Writing mattered to him because it was a forcing function for rigorous thinking about important decisions. As one Amazon veteran told Coulter, writing the memo forced them to confront their own ideas and clarify what they wanted to say. And this often meant that writers would change their mind several times while writing the narrative.
It is true that building meetings around memos disciplined the readers as well. Because meetings began in silence, everyone worked through the same argument before anyone spoke, and the discussion started from the substance. But if the content of a memo carries less of the writer’s thought, the value of reading it diminishes in proportion to that fall. Time spent studiously reading well-written but empty texts is time wasted.
The underlying problem is that AI can break the link between thinking and the production of text. Steve Huynh, a former Amazon principal engineer, recently described a review meeting in which a product manager began discussing a six-page memo he was supposed to have written. Almost immediately, it became clear that he was reading it for the first time. He was presenting something that he had not thought about at all.
Readers are adapting to the change in the writing process. Coulter reports that one Amazon manager now asks for longer documents because he plans to have AI summarize them, and Huynh says some employees now tag suspect memos “AI;DR”: AI, didn’t read. In a Stanford and BetterUp Labs survey on workslop, 42% of workers who had received low-effort AI-generated work said they saw the sender as less trustworthy. Once a written product stops being evidence that someone thought about the content, readers simply stop giving it their attention.
The memo’s 3 jobs
The trap for leaders is to treat this as a problem about memos. Ultimately, businesses don’t need memos. They need good decisions. The memo mattered only because, for a long time, it reliably communicated the results of the thinking process that lay behind the writing.
Baseball provides a useful analogy here. In his book Moneyball, Michael Lewis tells the story of how the Oakland A’s responded to the loss of a star player, Jason Giambi, to free agency. Instead of trying to replace him with another player of the same caliber, the A’s realized they didn’t need another Giambi, per se. What they needed was the measurable impact he had on their results. So rather than search for one player to fill the gap, they tried to recreate him “in the aggregate” by finding several cheaper players who together delivered the same statistical impact.
There is a lesson here for how we think about Amazon’s memos—and, indeed, for how we think of much else in the new AI-powered workplace. We shouldn’t be looking for simple substitutions of AI work for human work. Instead, we should look at how work produces value in aggregate and then redesign workflows to reproduce the end result, not the individual steps.
A well-written memo bundles three distinct jobs into one artifact. It forces authors to think, it provides evidence that the author has done the thinking, and it gives its readers a shared understanding of the issue at hand. What the Oakland A’s can teach us is that, rather than trying to restore all of those functions in one stroke, the better path is to make two distinct moves: one that restores the first two jobs together, and one for the third.
Force the thinking, then prove it
The rebuilding process starts with honesty about where thinking is actually needed. A great deal of human writing does not provide the basis for making important decisions, so in those cases there is little reason to build a replacement process that forces the thinking and evidential functions alongside the communicative role of the writing. A report summing up project progress over the past week, for instance, can simply be an assemblage of information, and the process of collating and communicating it can happily be automated. But when it comes to visions of the future or arguments about which decisions are best, leaders have to adapt to the fact that polished prose no longer guarantees deep thinking.
The first move is to both rebuild the forcing function and provide evidence of the underlying thinking. This can be done by making the author, rather than the document, the thing that gets examined. Amazon’s meetings already included the seed of this requirement. Once the silent reading ended, the writer had to answer for every line. The change for the AI era is that the defense becomes the initial gate. A memo earns its place on the agenda only if its author can walk the room through the argument and take questions without the document in front of them. That is not something the author can hand to Cedric, because Cedric will not be in the room, and knowing the defense is coming does what the blank page used to do.
Terence Tao, the Fields medalist, has proposed the same rule for mathematicians: If the authors of an AI-assisted result cannot give a clear, expert-level talk on it, the result should not be published. A proof that no human can properly explain, he writes, “should be viewed as incomplete, even if it has been formally verified.” The same holds for a memo.
Who does the understanding?
The memo’s third job, helping create shared understanding, seems the least problematic, but it raises a counterintuitive challenge. We tend to think that AI-written texts are problematic because they waste our time by spreading poorly thought-out ideas. Restoring the underlying thinking should, then, automatically solve the communication problem as well. But things are not quite so clear cut. Even when the content is genuinely good, AI can help humans produce more valuable outputs than other humans can absorb.
We can see this in the challenges that the discipline of mathematics is facing. Mathematicians are being overrun by journal articles and preprints containing AI-generated proofs that nobody can properly digest. In September, OpenAI said its models had solved one of math’s Millennium Prize problems. A month later, it released hundreds of AI-generated results in one batch. Many of the proofs have been verified by software that checks their logic, but knowing that a result is correct is not the same as understanding why it is. The Association for Human Mathematics put it bluntly in a statement reposted by Tao: “We reject OpenAI’s assertion that this release advances our subject.”
Pavel Etingof, a mathematician at MIT, argues that AI’s results truly become part of mathematics only when a human reads them, understands them, and gets something out of them. A thousand new proofs sitting unread on a shelf add nothing to the discipline. The same logic holds for companies grappling with a surfeit of good writing outputs; no matter how intelligent a memo is, it becomes a part of a company’s decision-making only if the relevant humans absorb it. That points to a new division of labor: AI can increasingly shoulder the generative burden, but the understanding still has to be done by people.
That makes human understanding a scarce and valuable resource, and leaders may have to start rationing it. Working out how to filter and direct the flow of outputs becomes an entirely new type of task. The preprint server arXiv, for scientific papers, now limits authors to two submissions a month. A company can do something similar, such as limiting the number of memos that get a silent-reading slot in a given month.
The artifact is not the thinking
For two decades, a six-page memo was reasonable evidence that someone had thought hard. AI has broken that link, and no policy will fully restore it. What leaders can do is stop mistaking the document for what it was meant to produce. The alternative is more six-pagers than ever, and fewer people who understand what’s in them.
It is difficult to open any newspaper (I know, who still does that anyway?) or consume any news, let alone join any business discussions, without having to talk about AI. The topic does not go away, though within this broad topic there are so many layers and subthemes and messages that it rarely gets boring, anyway.
As Garrison Lovely noted in his brilliant book, Obsolete, people tend to fall into three categories when it comes to AI attitudes. There are the accelerators, who are basically in the utopian camp and accordingly convinced that artificial intelligence will solve all of our problems, so we cannot go fast enough in our efforts to accelerate and evolve AI. At the opposite extreme, we can find the worriers, who think AI-induced human extinction is looming and even if not, there’s a strong chance that AI will have devastating effects on work and life. And then you have the critics who are broadly skeptical about most popular claims about AI, whether positive or negative, and perhaps most controversially believe that AI is simply overrated.
Since nobody has data on the future, we will avoid picking camps here—time will tell, and by “time” we should really mean the next two or three decades rather than 2027 or 2028. Indeed, obsessing over the future can be an unhealthy strategy for not dealing with present problems and concerns, including how to remain relevant in an age in which AI is unquestionably performing a large number of “knowledge economy” tasks that were hitherto the exclusive remit of human intelligence. As I predicted in I, Human, AI’s growing ability to compete with human skills is not just redefining what it means to be human, but also forcing us to double down on the things AI cannot replace even in order to remain employable.
Key indicators
So, how should you assess your employability and career success in today’s AI age, and what are some positive signs that you are unlikely to be automated or made obsolete by AI? As of today, here are five critical indicators to consider:
(1) You have a “messy job”: As Luis Garicano at the London School of Economics wisely noted, one of the best ways to resist automation is to have a chaotic, unpredictable, or “messy” job. Indeed, AI’s value is mostly in automating tasks, but you can only automate what you can standardize first; conversely, if you automate chaos you get chaos at scale. This is not to be mistaken for having a disorganized or chaotic mindset, which makes even predictable and orderly things seem chaotic, but rather an objectively unpredictable, complex, and nonrepetitive working routine—or if you prefer, a nonroutine!.
(2) You have a “hungry mind”: As my colleagues and I showed, contemporary careers require not just a strong work ethic and people skills, but also high levels of intellectual curiosity. Indeed, this is even more pressing in the AI age, which removes incentives for developing or harnessing human curiosity—why learn, after all, if you can answer any question at your fingertips just by prompting genAI? In that sense, you can think of genAI as the intellectual equivalent of the fast food industry. It’s there 24/7 and provides you with a tasty, quick, cheap but not very nutritious fix, and it can be quite addictive too. Just like our hunter-gatherer ancestors didn’t need standing desks, gym memberships, or Pilates (for they were active all day and only ate very healthy and nutritious food . . . IF they were active enough to procure it!) it is conceivable that their offspring may need to develop new cultural adaptations in order to keep their mental muscles from atrophying. For example, they may go on “thinking retreats” to stimulate those parts of the brain that have become nearly dormant because every thinking task is outsourced to AI. Until we get there, curiosity will keep some people more active in the thinking and learning departments, and for them there is hope. (Incidentally, this will also allow them to understand and leverage AI more, including for augmentation purposes.)
(3) You are entrepreneurial: While employers are trying to figure out what AI will do to their organizations and jobs, some people are finding ways to leverage AI’s disruptive and innovative potential. Not so much to delegate boring and tedious tasks—typically, the value AI brings is not in what it automates, but rather what humans learn to do with the time they save—but in order to experiment and test new ideas and innovations. In essence, entrepreneurial people see opportunities where others don’t, and they are quicker than others to act on those potential opportunities by bridging the gap between demand and supply. For example, in the AI age, a single individual can now launch a business that would previously have required a small army of employees, using AI to conduct market research, develop products, build websites, create marketing campaigns, manage customer relationships, and even provide customer support. Likewise, professionals can turn their expertise into new products and services, identify unmet customer needs, or experiment with business models that would have been too expensive or time-consuming to test before. In that sense, AI is not just a tool for doing existing things faster or cheaper, but an opportunity to do things that were previously impossible, or at least economically unviable. The critical difference, of course, is that while AI makes it easier than ever to execute an idea, it does not necessarily make people better at coming up with good ideas in the first place, let alone identifying which ones are worth pursuing.
(4) Your employer is seriously committed to “augmentation” rather than “automation”: Most organizations promise their employees “augmentation dreams,” which roughly translates into a utopian world where their jobs are more interesting, fun, and rewarding, and they are more productive because they have AI as their copilot or co-intelligence. However, many of these employers give up on the vision and instead yield to the temptation to automate and cut entry level jobs just to show some ROI from AI. Fortunately, some employers have not given up yet. You may consider yourself fortunate and less vulnerable to automation if your employer is genuinely trying to figure out how to augment rather than automate humans in the AI age. As always, choosing your employer (and boss) wisely is often more consequential for your career than whatever skills, personality, or talent you may have.
(5) Artificial general intelligence is less likely than techno-enthusiasts think: As Garrison Lovely explains in Obsolete, much of the debate about AI’s impact on humanity rests on assumptions about the future that are presented as if they were facts. In particular, the notion that AI will inevitably progress from today’s impressive but imperfect systems to artificial general intelligence (AGI), capable of matching or surpassing humans across virtually every cognitive domain, remains a prediction rather than an established scientific fact. Indeed, there is a rather significant difference between building machines that can perform an ever-growing number of intellectual tasks and creating machines that can think, reason, learn, and adapt as flexibly as humans do, particularly in unfamiliar situations. Of course, this does not mean AGI is impossible, or that we should dismiss the very real risks of AI simply because some of its most enthusiastic advocates have a tendency to confuse science with science fiction. Nor does it mean that AGI is necessary for AI to displace millions of workers. However, the assumption that human intelligence is merely a temporary competitive advantage, destined to disappear as soon as machines become sufficiently powerful, is far from proven. In that sense, perhaps the most reassuring sign that you will not become obsolete is that nobody, including those who are most confident about predicting our obsolescence, actually knows when, whether, or how AGI will arrive. And while betting your career on the failure of technological progress would be unwise, betting against the accuracy of Silicon Valley’s predictions may be a rather more sensible strategy.
Ultimately, the best way to avoid becoming obsolete is not to obsess over what AI may or may not be able to do in the future, but to focus on what you can do today to remain valuable, relevant, and adaptable. This means cultivating the skills, mindsets, and opportunities that enable you to work with AI rather than compete against it, while remembering that your employer’s choices may matter as much as your own. Even if AI stopped evolving today (which, let’s face it, is unlikely), we would still need to work hard to upgrade ourselves as humans, especially to be future-fit for an employment market that has already been shaken up by AI.
Six months ago, the outcome of the November 3 midterm elections would have seemed impossible. Most states and regions were highly polarized politically. So-called purple states were generally a reflection of a balance of voters from different areas of…
While Democrats’ legislative agenda for the incoming Congress is still taking shape, early indications suggest workplace and labor issues will have a prominent role.
And while the Senate filibuster and President Donald Trump’s veto power make it un…
Whether you’re among the 53.5% of Americans who get their health insurance through work, the 20.1% enrolled in Medicare, or the 10.5% purchasing ACA coverage, mid-October or early November typically marks the start of the open enrollment period for ne…
In the aftermath of the Democrats sweeping the midterms, the internet has anointed this moment “Woke 3.0”: an extension of the terminology that people have used to describe the political and cultural movement that took hold during the pandemic, partic…
Hiring in October blew past expectations, with U.S. employers adding 165,000 jobs and unemployment edging down to 4.1%.
The Bureau of Labor Statistics (BLS) also revised its jobs estimates upward for the previous two months, concluding that the eco…
I spent the first half of my career as the product manager at a software company. Product teams are constantly reviewing their products, assessing: What do our customers want? What changes should we make? Do we (maybe) need to retire a product altogether because it no longer serves its purpose?
Most solopreneurs don’t take the same approach. They continue to push the same products or services, or tack on new items until they have a large suite of offers they have to maintain. We rarely take anything away.
I’ve gotten into the habit of an annual review to make sure everything I put out into the world for my business still makes sense for both me and my customers.
The trap of adding and never subtracting
Part of the joy of running a solo business is launching something new. You get to make all the decisions about what an offer looks like and how to launch it.
Removing an offer is a lot less fun. It can feel like a loss. Plus, you may have material (website, slide decks, email sequences) that need to be updated.
The problem is that every offer needs upkeep. Content gets out of date. You may also be paying for tools related to hosting or delivering an offer that isn’t doing anything for you.
I create a lot of resources for solopreneurs, often describing systems or processes that make a solo business run efficiently. These can quickly feel dated—or even obsolete—as technology changes.
For a long time, I spent my precious hours creating new resources without a process for updating existing material. That’s a mistake, since outdated information quickly erodes trust.
The questions I ask about every offer
Now I take time to review every product and service I offer and ask myself the same questions:
Is anyone buying it? I start with sales or bookings from the past year. Low numbers aren’t automatically a reason to cut, but I need to review the offer more closely.
Do I still want to do this work? A service can sell well and still be work you don’t enjoy.
What does it cost me to maintain? I count the time I spend updating information and supporting it, plus the time it takes to promote. Then I add any tool subscriptions tied to it.
Does it still fit where my business is headed? An offer that made sense two years ago may not match the clients or work I want now.
I maintain a business road map and value statements, and compare my offers against these documents. I use AI to help this process by having Claude do the initial comparison. This becomes my checklist for a human review and assessment.
What to do with an offer that isn’t working
Keeping an offer as is and cutting it aren’t the only two choices. There are a few other paths:
- Refresh it: Update outdated content.
- Reprice or repackage it: Raise the price or bundle it with something else so it doesn’t stand on its own.
- Re-promote it: Sometimes there’s nothing wrong with the offer; perhaps it just hasn’t been marketed enough.
- Retire it: Take it down, and let existing customers or clients know if needed.
I have a resource related to AI tools for solopreneurs. It made sense when I first created it several years ago. Now it’s impossible to maintain because the AI changes at lightning speed. Instead of spending my time updating it, I spent my time finding all the places where I reference this resource and removing it.
But to start, the most important step is making the time for a review. Block off time on your calendar once a year or put it on your to-do list. The review process is separate from acting on your decision. Your goal is to land on “Where do I go from here?”
Tokenmaxxing is on the way out, even as some of the biggest players in the tech industry are urging their employees to use AI as much as possible. A new Bloomberg report reveals that Apple is cutting employee budgets for AI token usage while remaining…
