What Comes Next?

On anxiety, uncertainty, and all of this AI/professional angst.

Welcome to Young Money! If you’re new here, you can join the tens of thousands of subscribers receiving my essays each week by adding your email below.

---

Change is in the air, or so it seems.

Over the last three weeks, I’ve had more conversations with friends and acquaintances who are “considering different career moves” than I’d had in the last year up to that point. Part of this is, as it always has been, structural.

I’m 28 and coming up on two years out of business school, which means most of my peers are ~5-8 years into their careers, and my former classmates are now ~2 years into their “post-business-school” paths.

For the business school folks, it’s pretty common to recruit for some version of banking/consulting/private equity/whatever, knowing you’re signing up for a two year role where it’s either “up or out.” You work really hard to get that job, and the assumption is, likely, that around 12-18 months into the job, you will have a feel for 1) if you want to be up or if you want to be out, and 2) if “up” is even a possibility based on your performance and the likelihood of room/mobility at higher levels.

The issue, when you reach that 12-18 month point, is that, unless you are all-in on your current thing, you’re probably having second thoughts. “I could make VP, but should I? Is that the right move?” And that uncertainty yields anxiety.

Many of my non-business-school but otherwise ambitious-and-successful friends are having similar thoughts? “What comes next?” “What comes next?” “What comes next?”

The late-20s-to-early-30s window is primed for anxiety and uncertainty because it’s where clearly-marked paths evaporate. You can either commit to climbing an increasingly exclusive, difficult, but lucrative ladder that you’re currently on, or take a leap and try something new. But there’s no blueprint for “new.” The current ladder offers stronger visibility into your expected compensation and status. “New” might mean that you have to upskill in areas that are currently unfamiliar. It almost certainly means reduced “guaranteed” compensation in exchange for higher hypothetical upside. But most of us in the land of “white collar” labor spend our formative years optimizing for the “de-risked” option. The issue, once you’re 5-10 years into your career, is that the “de-risked” option quietly stops being the “de-risked” option as the definition of risk shifts.

At 22, the “risk” of not landing a premier job can mean a slower career trajectory and lower earnings. For all of the shit that “investment banking” gets, anyone who can land a job at Goldman out of undergrad almost-certainly should take it. You’ll make more money than you would anywhere else, and you also get a stamp of approval that screams “I’m competent.” Legible competency is quite useful in life (why else would I have gone to business school? It’s a $200,000 stamp of competency).

But that risk curve shifts if you spend too long on any particular path. Once you’ve stacked a few pedigree stamps (good schools, jobs, etc), the greater risk, I think, is failing to leverage that pedigree to explore new opportunities. The “risk,” if you don’t make a change, is that you find yourself investing 70+ hours a week into climbing an increasingly-narrow ladder, where burnout, fatigue, and career stagnation are real, while the folks who took a differentiated path at, say, 26, probably suffered a short-term compensation and status regression, but compounding then kicked in, and by 30, they were likely running laps around the hypothetical version of themselves that never deviated.

And the longer you stay on current path, the harder it is to deviate from current path, because inertia kicks in. Uncertainty is uncomfortable but not paralyzing in your 20s, but if you spend a decade doing one specific thing, and particularly if that “thing” is a well-defined, well-trodden career path, you’ll find yourself unable to begin conceptualizing what “something new” might look like. If you spend too long optimizing for certainty, your risk perception becomes fried.

The most important traits, I think, are 1) the ability to learn new skills, and 2) the confidence in your ability to learn new skills. Malleability is one of the only competitive advantages with staying power. Especially with AI blowing up the traditional model of how pretty much the entire white collar economy works.

And AI is blowing up the traditional model of how pretty much the entire white collar economy works. If your job has consisted of some version of “move things around on a computer,” it’s in the process of being totally commoditized, whether or not you realize it.

There was, basically since the invention of computers, some version of a “stable career path” that went something like this: earn your stripes doing the BS “analyst / junior” tasks for a few years, slowly take on more responsibility, and, eventually, you can become “the guy.” That path is disappearing, or, at least, certainly changing shape. At the company level, pure-play software businesses that offer specific point solutions are around a year away from being commoditized by AI model providers. At an individual level, much of the “work” of most junior-level white collar jobs can be “done” by AI now, and, more importantly, one junior employee who has actually been curious enough to figure out how to use Claude Code, or Codex, or Gemini, or Cursor, now has the leverage to handle the work loads of a team of 4 juniors previously. That’s the real shift: I don’t think AI “kills” all white collar work, but it does enable much leaner teams, which changes organizational structures and career paths.

I’ll give a more specific example from my workflows:

I now have my email, calendar, and Granola notes synced to Claude to 1) maintain project folders with full context on any startups I’m looking at so I can instantly query any information relevant to any particular deal, and 2) have AI send me an automated email each day recapping all meetings and laying out existing “to-do’s.” It’s still not perfect, but it probably saves me a cumulative 10+ hours per week of “looking for stuff.” If I spend a full weekend really going deep on current capabilities, I can probably double my time saved on “admin-ish” tasks. All of that freed up time means I can meet with more founders, or spend more time in deep research on different market sectors. AI, used well, creates leverage. And companies know this.

Goldman Sachs is trying automate back office and internal functions with AI wherever possible. Anthropic built Claude Cowork in ~10 days, and the whole thing was “built” by Claude Code.

Leverage. Leverage. Leverage.

Ever-improving AI models with ever-increasing context are going to be able to “do” anything on a computer, and I think that most people’s ideas of AI still being “chatbot that hallucinates” stem from lack of experience with what’s out there today. This stuff is nuts, and coding assistants are the leading indicator of what’s to come everywhere else. All software can be, and, at Anthropic, who I think is “winning” the AI game, is, written by AI today. That’s only going to accelerate.

A year from now, Claude for Excel is going to be able to ingest any source materials you provide it and spit out a model showing whatever you want, however you want, in minutes. You’ll be able to link your bank account, credit card statements, and pay stubs, and have AI spit out your taxes for 2026. Legal contract review will be totally commoditized.

This isn’t to say that every “business” is a zero, but most businesses are heavily bloated in a world where AI can operate increasingly autonomously. Tokens are cheaper than new hires, particularly when you don’t have to “train” the AI on anything. You just say, “Do thing,” and it does it.

Even if all “developments” in AI models stopped tomorrow, their current abilities are powerful enough to permanently reshape what labor markets look like as they continue to penetrate the economy. The world, quite literally, can’t keep pace with the speed of improvements in models, and it’s only accelerating.

Back to the career stuff.

Whatever model of the “professional” world that you had three years ago is almost-certainly outdated. Simply put, many ladders that were impervious over the last ~50 years have either been pulled up or disappeared entirely, and many of the things that did work will no longer work because things that used to be valued have now been commoditized. Pattern-matching off the prestige of the 2010s, or even early 2020s, is no longer a high-signal decision making process when considering “what you should do?” It’s not that the prior paths were Thanos-snapped out of existence overnight, but previous headcount needs in what were otherwise unassailable roles/paths/positions have just been greatly reduced.

And yet, at the same time, this massive wave of change brings with it opportunity. Littlefinger was right; chaos is a ladder.

So, what comes next?

I would say that there’s two ways to play the next few years. Both of which involve getting comfortable with the uncomfortable, particularly with leveraging AI. The first is that you become proficient with whatever AI offers, apply it to the existing world, and put yourself in a position to run laps around everyone else who is otherwise either failing to appreciate this shift or is too paralyzed to take advantage of it. AI doesn’t just give companies leverage to hire fewer employees, it gives employees leverage to have an outsized impact. You want to be the employee who has the leverage. LLMs are performance-enhancing drugs for people who use Excel and VSCode. Like a baseball player in the 90s, you’re crazy if you’re not using them to up your game.

The other opportunity is identifying which versions of the world are “outdated,” and molding yourself to the shape of what you believe comes next. This could mean starting your own company. It could mean pitching yourself as an “AI-first” version of a role within an organization that may not have previously existed. Either way, the only things that matter are identifying what got commoditized, avoiding putting yourself on a track headed for commoditization, and identifying how to best position yourself where the puck is going.

Ironically, the skills that were most valuable even a few years ago, i.e. playing the “career game,” are likely outdated now. Particularly when consulting firms face pricing pressure, investment banks are rapidly automating work, the time to make partner in big law keeps getting extended, and the music stopped for “easy money” in PE when interest rates were hiked.

Independent thought and curiosity now trade at a premium when the previous ravine-like gap between “idea” and “execution” has been reduced to a few inches. I don’t buy the narrative that “everything is cooked” because AI is getting really good, but I do think that the days of speed-running an existing status game to successfully are in their twilight phase.

Put differently, you can no longer expect any particular job to “save you” of the task of figuring out what to do with your life. 20 years ago, you could outsource your sense of agency to whatever your alumni network was doing and probably make a lot of money along the way. I’m not so sure that’s true any more with how fast everything’s changing. But, conversely, it’s never been more possible to dive headfirst into something new. You just need the wherewithal to identify where those opportunities might be, and the willingness to explore them.

My personal opinion? The rapid re-organization of all things “status” is probably a good thing. At least now we can stress about achieving our own goals rather than those bestowed upon us by others.

Anyway, happy Thursday.

- Jack

I appreciate reader feedback, so if you enjoyed today’s piece, let me know with a like or comment at the bottom of this page!