How To Get Hired During the AI Apocalypse

And other discussions to have with your kids on the way to college graduation

May is a season of rituals and milestones. Many of my friends headed out this weekend to engage in perhaps the proudest parent spring ritual : the college graduation ceremony.

One friend heading to Notre Dame had her nails painted bright green to support her daughter’s soon to be alma mater, the “Fighting Irish.” It’s an exciting time for any family, the precipice of adulthood, the end of the K-12-college structured gauntlet. Now, anything could happen.

Or not.

I want to feel happy for new grads and the parents who helped to fund them, but I bite my lip. Most of my friends' kids don’t have jobs lined up. And we’re instructed not to ask.

Photo of my kids when one of them graduated from Temple University.

It’s been 3 months since my book “Who Needs College, Anymore?” came out and I am anxiously (not in a good way) watching my predictions come true faster than I expected. At first, as I wrote in March, colleges felt under threat from the Trump Administration, with the blocking of research funding and gutting of the Education Department. But now as I continue a flurry of book events, conference appearances and zoom panels into the spring, meeting with college leaders, accreditors, businesses, parents, a different set of headline harbingers is dominating the conversations. A less temporary, more existential threat to the four year degree: AI could hollow out the entry level job market for knowledge workers (i.e. new college grads). And if 56% of families were saying college “wasn’t worth it” in 2023,(WSJ), what will that number look like in 2026 or beyond? The one of my kids who went to college ended up working in a bike shop for a year-ish after graduation. No regrets, but it came as a shock to them that they weren’t more employable with their neuroscience degree.

Derek Thompson just wrote a piece in the Atlantic that got folks fired up on LinkedIn, featuring a scary chart suggesting that the unemployment rate for new college grads is 5.8%, which is worse than the overall rate. Presumably, this means that you have a better chance of getting a job without a college degree than with one. Thompson gave three reasons that might be at play, a soft economy, employers de-emphasizing degrees for hiring and AI. Like my fellow observers, I hopped on the AI hand wringing bandwagon. Because it began to validate the anecdotal evidence I was gathering from employers I interviewed for the book. Thompson wrote: “You might suspect that…

“As law firms leaned on AI for more paralegal work, and consulting firms realized that five 22-year-olds with ChatGPT could do the work of 20 recent grads, and tech firms turned over their software programming to a handful of superstars working with AI co-pilots, the entry level of America’s white-collar economy would contract.”

We are in the eye of the cyclone with respect to AI’s impact. We can’t see the future very clearly from here. But colleagues talk about the post-Artificial General Intelligence (AGI) landscape as if they were imagining a world after an apocalypse.

I was in the eye of the cyclone when the digital world first came of age, working to build consumer products on that new thing called the Internet. So this moment feels familiar, but not. As my friend Larry Roth, who worked with me at AOL in the 1990s mused about the difference: “Then, we were imagining how to build applications on a new platform. Now, we are bracing for the applications to build themselves and we wonder where they will take us.”

The loss of control is obviously the nightmare scenario. But those of us focused on education-to-work are looking more practically to get a handle on where the AI technology evolution will become friend or foe.

An early sampling based on my conversations:

So, on the book tour, I have been asking everyone I meet at the intersection of learning and work: What should we be designing to help learners and early earners through this interstitial period? Optimists will tell you that the competency that will survive the apocalypse is “learning how to learn” and colleges should make that the backbone skill of their degrees. And figure out a way to specifically credential it. But my friend Ryan Craig wrote in his newsletter this week, “Isn’t it more important to learn how to DO?” That’s what employers really care about. What are we going to DO with what AI can learn for us?

A colleague provided a great example: Her son, newly graduated, went for a job interview as an entry level writer last month and he was asked, as a test, to produce a story with AI and then use that story to write a better one by himself. He would presumably be judged on his ability to prompt AI and then improve upon its product. Is that learning how to DO? I think so. It’s using AI tools to accomplish a workplace task.

And that’s one way I am thinking about how to teach AI skills for now. To contextualize them in workplace tasks, real or simulated, at least in this period while we are waiting for the inhuman intelligence to reach its humiliating zenith.

I am considering devoting my design thinking course that I teach at George Mason University to this design topic next year, where I would work with the students to define and “credential” what “AI ready“ might mean across their different majors. (Message me if you have suggestions.)

For now, I am taking a look at the competencies that we at the Education Design Lab created a decade ago, when employers were complaining that learners didn’t have enough “soft” skills, what we now try to call durable or human skills.

Here is the subway map that charts the 9 micro-credentials we created in 2014-2015. Each micro-credential could be earned by passing an assessment on the four sub-competencies, (some of which overlap).

Above: Education Design Lab’s Durable Skills Framework

When you look at the competencies that make up, say, critical thinking, it feels like those skill sets are doubly important in the age of AI:

Gather and Assess Relevant Information,

Identify Patterns,

Draw Conclusions,

Question Assumptions.

Obviously we need to sprinkle a healthy dose of tech literacy in here. And several of my employer friends are mentioning “agentic project management” as a new management skill. Which means: can you manage the output of multiple bots towards a desired task or challenge?

Could we put some of these skills together and form an “AI ready” credential. I think I believe that credential will only serve the newly homogenized applicants if they can stand out with some actual job experience.

Tyler Cowen and Avital Balwit remind us not to think too narrowly in a chilling new article called “AI Will Challenge What it Is To Be Human. Are We Ready? (Free Press subscription required)—They write, “We stand at the threshold of perhaps the most profound identity crisis humanity has ever faced.” And the biggest thing they worry about is will we feel demoralized rather than freed or even energized by robot agents. They give a lot of great examples across all parts of our lives of how we might fight back. I will stick to how to maximize the journey from school to work to set ourselves up for lives of meaning and economic success.

In my book, I call for a redesign of college institutions to make way for these changes. But how would I retrofit an existing degree structure in the meantime to the post-AGI apocalypse that is coming? Well first, I should ask AI: How would colleges help their grads get hired in an AI age?

ChatGPT is upbeat and gives a shout out to its own competitors:

…AI won’t replace most knowledge workers—but knowledge workers who use AI will replace those who don’t. The winners will be:

+Fast adopters of AI tools (e.g., Copilot, ChatGPT, Tableau, Notion AI)

+People who shift from routine output to judgment, communication, and creativity

+Those who upskill continually in data literacy, AI prompting, and interdisciplinary thinking

OK, now I need to take ChatGPT’s answer and build on it. Apparently originality (and charisma, according to Cowen and Balwit) will be key AI-proof skills. My interim idea:

I have been telling college students to major in journalism. Not that there are many jobs in journalism after the last tech revolution. I left journalism 20 years ago. The sector lives in the shadow of its golden age (when I was lucky enough to be a reporter.) But the skills you learn as a journalist are a decent proxy for all knowledge workers at this crossroads. You are called upon to be agile, curious, critical, to show initiative. To synthesize information, find patterns, question assumptions, but also, importantly, to put out a product. You are called on to DO something and produce your pieces on deadline. Every employer loves that. And, I believe, always will.