The Academy — Mar 13, 2026

The learning-to-earning pipeline had a clarifying week: the institutions and employers actually building plumbing — stackable credentials, guaranteed transfer agreements, paid apprenticeships, mandatory AI fluency — are pulling away from those still workshopping strategy decks. Community colleges are quietly becoming the growth engine of American higher education, federal hiring just codified skills-over-degrees into real operating rules, and Accenture's CEO compared requiring AI proficiency to requiring employees to use a computer in the 1990s. The gap between organizations that are moving and organizations that are meeting about moving got wider.

If you've been running on the assumption that the demographic cliff plus FAFSA chaos equals permanent enrollment winter, the National Student Clearinghouse just updated your priors. Its final Fall 2025 report shows total U.S. postsecondary enrollment up 1.0% year over year, driven almost entirely by undergraduate growth at community colleges and public four-year universities. Private nonprofits fell 1.6% year over year. For-profits dropped 2.0% year over year.

The structural story underneath is even more interesting. As of Fall 2025, nearly 500,000 students enrolled at four-year institutions had previously started at two-year colleges. Students 17 and under — the dual-enrollment cohort — accounted for roughly 28% of the year-over-year undergraduate enrollment increase. Dual enrollment grew about 10% year over year. This isn't a blip; it's a reordering of how Americans enter higher education.

A concrete example of the stackable-pathway play landed this week too: Central New Mexico Community College signed a transfer agreement with the University of Wisconsin–Stout, letting CNM's Industrial Automation certificate grads move directly into an online bachelor's in advanced manufacturing. That's a local, hands-on credential turned into a clear route for upward mobility — the kind of low-friction articulation that makes a press release into an actual pipeline.

Meanwhile, the Clearinghouse's sixth annual Some College, No Credential report shows over one million adults who previously stopped out have returned to higher education, with many enrolling in shorter, stackable programs rather than full degrees. Adults with about two years of prior credit have the highest completion odds. For institutions still counting "adult learners" as a monolith rather than targeting the sub-segments with demonstrated completion and wage-gain potential, the data is now specific enough to build a strategy around — or to expose the absence of one.

Most companies say they're reskilling for AI. Accenture is the rare one that made it non-negotiable — and the CEO said so out loud.

In a recent podcast, CEO Julie Sweet said AI proficiency is now mandatory for working at the firm and moving up its ranks. The company invested more than $865 million in a six-month optimization program, reskilling thousands of employees and showing the door to those who refused to adapt. It's part of a three-year, $3 billion push to integrate AI, with a goal of doubling AI talent to 80,000 professionals across a workforce of more than 770,000.

Sweet's framing is pointed: she compared requiring AI fluency to requiring employees to use a computer in the 1990s. Not coercion — just how a modern firm operates. The contrast with the broader market is sharp. As of Q4 2025, only 38% of companies reported integrating AI to improve workplace productivity, per Gallup — a 1 percentage-point increase quarter over quarter.

Meanwhile, Walmart is scaling its own internal pipeline with a Digital Skills Academy aimed at training 100,000 associates in data analytics and e-commerce over three years; early internal data suggests roughly 30% of participants move into higher-paying roles within the program's first year. The pattern: big employers are building their own training infrastructure rather than waiting for institutions to sell them credentials. Whether that's a complement or a competitor to higher ed depends entirely on whether colleges can demonstrate employer-aligned outcomes faster than employers can build their own academies.

If you want a preview of what a serious local talent strategy looks like, watch Colorado's Pikes Peak region. Districts are guaranteeing tuition at Pikes Peak State College, layering on "Colorado Promise" last-dollar aid for families under $90,000, and expanding concurrent enrollment so high schoolers earn college credits at community-college price points. As of 2025, more than 40,000 Colorado students are taking community college courses while in high school, up 3,400 year over year. Early college high schools in the region require students to complete at least 60 college credits or an associate degree before graduation.

The workforce side isn't an afterthought: Innovate Pikes Peak pools urban and rural district resources for in-demand fields, while new facilities like the Colorado Springs School of Technology focus on aerospace and cybersecurity credentials, internships, and mentorships. The initial rollout expects over 500 participating students in year one with roughly 30 local business partners, and districts are leveraging Perkins V funds (the main federal career-tech funding stream) to scale employer-connected placements.

Buried in the coverage is a statewide target worth flagging: Colorado wants 100% of the Class of 2029 to graduate with quality work-based learning, college credit, or an industry credential. That effectively treats CTE experiences as a graduation norm, not an elective — and will reshape how districts allocate staff, schedule time, and prioritize employer partnerships across the state.

Your CEO may be bragging about dropping degree requirements, but a new University of Phoenix Career Institute report suggests very little has changed in the actual hiring funnel. The latest Career Optimism Special Report — subtitled The Illusion of Progress in Skills-Based Hiring — finds that while many employers claim to embrace skills-first approaches, they lack the standards, tools, and processes to evaluate skills consistently.

The problems are familiar to anyone who's watched this space: job descriptions still default to degrees, applicant tracking systems can't parse non-traditional credentials, and interviewers revert to pedigree as a proxy for competence. Jobs for the Future's Alison Lands is blunt: skills-based hiring only moves the needle when "intent and infrastructure are aligned."

This week's federal news sharpens the contrast. The 2026 Merit Hiring Plan includes measures to codify skills-based hiring for federal roles, lowering the long-standing "paper ceiling" around four-year degrees across job series like program analysis and IT. If agencies succeed in embedding this into day-to-day hiring, federal jobs could become the largest live testbed for skills-first recruiting at scale — with ripple effects on how every employer interprets non-degree credentials.

The D2L synthesis of employer training statistics (2025) puts numbers on the execution gap: it reports 85% of employers say reskilling is a priority, but the infrastructure to make training stick — assessment design, manager upskilling, integrated HR systems — remains incomplete. For L&D leaders and edtech founders, the opportunity is in the unsexy middle layer: skills taxonomies, assessment frameworks, and recruiter enablement, not yet another badge.

A Federal Reserve Bank of Dallas economist has built one of the more useful frameworks for understanding what AI is actually doing to wages — and it's not the story most outlets are running. The key distinction: codifiable knowledge (what you learn from textbooks) versus tacit knowledge (what you learn by doing). AI can replicate the former but not the latter, which means it automates jobs that rely on codifiable knowledge while complementing workers whose value comes from experience.

The data backs this up with uncomfortable specificity. Wages are rising in AI-exposed occupations that place a high value on tacit knowledge and experience. But for new entrants, the job market in AI-exposed fields is getting very tough. New college graduate unemployment is trending toward 10% as of early 2026 — the highest since the pandemic recovery peak in July 2021. An analysis published by CNBC found a 50% decline in new role starts by people with less than one year of post-graduate experience between 2019 and 2024, consistent across sales, marketing, engineering, HR, operations, design, finance, and legal.

MIT research published this week (March 2026) adds a concrete mechanism: developers using GitHub Copilot increased time on core coding by over 12% while cutting project management tasks by nearly 25% and reducing peer collaboration. AI co-pilots don't just speed tasks — they reallocate what parts of a job are done by humans versus machines. That reallocation favors workers who can supervise, validate, and integrate AI outputs. The career ladder isn't just harder to climb — the first rung is being removed across multiple industries simultaneously.

The implication for curriculum designers is almost too direct to ignore: if your program produces graduates with lots of codifiable knowledge and no demonstrated tacit knowledge — no portfolio, no real projects, no supervised practice in messy situations — you're sending people into the half of the market that AI is actively eating. This is from a Federal Reserve economist, not a vendor white paper.

Accenture's CEO comparing AI fluency to learning to type in the '90s, BlackRock writing a nine-figure check for plumber school, and a Federal Reserve economist calmly explaining that the first rung of the career ladder is being sawed off while universities debate whether to add a portfolio requirement.

Somewhere a college is spending $2 million on AI integration, watching enrollment stay flat, and getting its bond rating downgraded — while a community college in Austin gets a foundation grant to have AI route alerts to actual humans.

Until next week. —The Academy

If someone you know is building pipelines instead of slide decks, they should probably be reading this. Forward it.