Over the past quarter, the same conversation has come up, independently, with multiple partners across our network. The headline isn't that AI education is growing. That's old news. The headline is that the shape of the demand is changing.
Learners and corporate buyers are no longer asking for basic "AI training." They're asking for more specific training, such as:
- AI for sales teams
- AI for marketers
- AI for analysts
The outcome of this new mindset: The job title is now in the request.
Let me be clear about what this is not: it's not a story about generalist AI programs fading. Foundational AI training is still the volume driver. Generative AI is the fastest-growing category on Coursera, with enrollments up 195% year over year, and AI literacy was LinkedIn's fastest-rising skill in 2025. Across our own partner network, foundational AI programs remain the highest-enrollment offerings, and nothing in the data suggests that changes soon.
The story is that a second layer of demand is forming on top of the first one. The first wave of professional AI education was generalist: "AI for everyone." The second wave is role-specific. The universities that build both layers and connect them will own what comes next.
What's actually happening in the market
The generalist wave is maturing, not shrinking.
"AI for everyone" certificates served a real need, and they still do. Most professionals need AI literacy first, or a baseline understanding of what the tools are and what they can do. Eight-eight percent of organizations now use AI in at least one business function, which means the baseline question has been answered at an enormous scale.
What's changed is the question that comes after the baseline. Learners completing foundational programs are asking the obvious next question: "How does this apply to my job, specifically?"
And employers are asking about it with them. CompTIA's workforce research found that the top format for AI education among employers today is training specific to individual job roles. McKinsey now lists role-based AI capability training as a core practice for organizations trying to scale AI rather than pilot it.
The foundational layer built the audience. The role-specific layer is where that audience is heading next.
The demand is forming around three categories first.
Sales, marketing, and data analysis are surfacing earliest across our network. Each has a distinct driver, and they're worth treating as three separate stories rather than a single trend with three examples.
- Sales has the clearest math problem AI is being asked to solve. Salesforce's State of Sales research found reps spend only about 40% of their time actually selling, with the rest going to prospect research, data entry, quote creation, and administrative follow-up. That's exactly where AI lands: Fifty-five percent of sales professionals now use AI for prospecting, and sellers who work with AI tools are 3.7x more likely to hit quota. But the questions sales teams bring us aren't generic. They're about call preparation, pipeline hygiene, personalized outreach at scale, and how AI fits within the CRM they already use. A prompt-engineering course doesn't answer any of that.
- Marketing has the loudest skills gap of the three. Ninety-two percent of marketers say AI has already changed their roles. CMOs are now putting 15.3% of their budgets toward AI, yet only 30% describe their organizations as AI-ready, and a lack of internal AI expertise is the single most-cited barrier. The gap between adoption and competence is stark: Only 17% of marketing leaders say their teams have received comprehensive AI training, while 58% name skills gaps as their top challenge. And the demand is segmenting inside marketing itself; what a content marketer needs from AI looks nothing like what a performance marketer or a marketing ops lead needs.
- Data analysis has the strongest labor-market tailwind. The World Economic Forum ranks AI and big data as the single fastest-growing skill category through 2030. Among data practitioners, AI is now part of daily work for 80%, up from 30% just a year earlier. And the audience is wider than data teams: Nearly half of companies are adding AI-focused analyst roles, and the managers who consume analysis, not just produce it, increasingly need to know how to question, verify, and communicate AI-assisted findings. That's a distinctly different curriculum than "Intro to AI."
The demand isn't coming from learners alone.
As we described in our last post, the team training inquiries that follow individual program completions are increasingly role-scoped.
What we're hearing:
- "We need this for our SDR team."
- "We need this for marketing."
- "We need this for our analysts and the managers who rely on their reports."
SHRM's research on L&D priorities points in the same direction: Role-specific training and personalized learning paths are the top focus areas, and workers dissatisfied with AI upskilling most often cite limited relevance to their actual role. The vertical request is showing up on both sides of the market.
Why generic isn't enough as the second course
Three reasons why role-specific training is becoming the follow-on layer:
- Generic AI training answers a literacy question. Role-specific training answers an applied-practice question. Once professionals understand what AI is, they want to know how to use it in their specific job. A generalist program can't answer that without becoming surface-level across every domain. A seller needs to build a repeatable prospecting and follow-up system. A marketer needs to produce campaign work that holds up to brand and performance standards. An analyst needs to know when to trust an AI-generated query and when to check its math. These aren't variations on a theme; they're distinct curricula.
- The work product looks different by role. A sales professional's AI use cases center on research, outreach, and pipeline management. A marketer's center on content production, campaign optimization, and analytics. An analyst's focus is on querying, verification, and data storytelling. Asking one course to serve all three meaningfully isn't a curriculum challenge, it's a structural impossibility.
- The buying decision is changing. Individual learners and corporate buyers can now evaluate AI training against a specific outcome ("Did this make our sales team faster at prospecting?"), instead of an abstract literacy goal. That raises the bar. Generic training can increasingly fail to clear it as the second purchase, even when it was exactly right the first time.
💡 The core shift: Professionals want to see their job title in the course name. That's not branding. It's a signal that the work has changed.
What this means for universities
Here are four observations from what we're seeing across the partner network:
- Generalist programs aren't the casualty. They're the foundation. This is worth stating plainly, because the data backs it: Foundational AI courses remain the most-enrolled AI offerings everywhere we look, from Coursera's top GenAI courses to our own partner portfolios. The strategic shift isn't away from generalist programs; it's toward stackable models where the generalist certificate becomes the on-ramp that feeds role-specific tracks. LinkedIn and Microsoft have already productized exactly this structure by offering 150+ AI certificates organized by role and level, with a foundational tier feeding the specialized ones.
- The build decision is harder than it looks. Role-specific programs require practitioner depth that generalist programs don't. An AI for sales program needs instructors who have actually carried a quota and understand AI, a meaningfully smaller pool than either group alone. Same for marketing, same for analytics. The institutions that move first will build the practitioner-instructor relationships that are genuinely difficult to replicate later.
- The corporate conversation is changing shape. Organizations that used to ask for "AI training for our team" are now asking for "AI training for our sales org" or "AI training for our marketing department." That's a different motion, closer to custom cohort programs than open enrollment. Institutions need to be set up for both.
- Sequencing matters more than coverage. No institution can launch eight vertical AI programs simultaneously. The ones winning the second wave are picking two or three verticals where demand is already visible in their inbound and going deep, not spreading thin across every possible role. The major online providers are moving fast here: Coursera and Microsoft have already shipped role-scoped AI certificates for sales, marketing, and data analysis. But the university advantage is depth: practitioner instruction, cohort structure, and institutional credibility that a self-paced certificate can't match.
How to think about sequencing your own portfolio
If you're deciding which role-specific programs to build next, three questions tend to cut through the complexity.
Where is corporate demand already showing up in your inbound?
The team training inquiries flowing in from program completers are a free-market signal. Look at which roles and functions those requests are scoped to. That's almost always your first vertical.
Where do your foundational completers work?
Your generalist AI program is already telling you what to build next, and the job titles of your completers are the demand map for your role-specific tracks. If a third of your foundational cohort works in sales and marketing, you don't need a consultant to tell you where the stackable demand is.
Where can you assemble a practitioner-instructor pool quickly?
Role-specific programs live or die by the depth of the practitioner bench. If you can't credibly staff instructors with real applied experience in the target role, the program won't meet the bar this audience is now setting.
Building both layers
The first wave of AI education was about whether professionals could engage with the technology at all. That question is being answered at scale, and the foundational programs answering it aren't going anywhere.
The second wave is about whether universities can meet professionals where they actually work: in a pipeline review, in a campaign sprint, in a data model.
The institutions that build for the job title, not just the technology, will define the next phase of professional AI education. The strongest position isn't choosing between the generalist program and the vertical ones. It's owning the connection between them.
If you're sequencing your role-specific AI portfolio for FY27, I’m happy to share what we're seeing across our partner network on demand signals and instructor availability. No form, no pitch.
AUTHOR
