How Venture Capitalists Built A For-Profit "Micro-University" Inside Our Public Flagships
You can graduate any time, but you can never leave.
Over the weekend, “prompt engineering” was briefly trending on Bluesky, as skeeters circulated two pieces of viral content, an alleged call for increasing scholastic “AI literacy” from NVIDIA CEO, Jensen Huang, and an Instagram ad promising certification in “AI prompting” through the Wisconsin School of Business’s Center for Professional and Executive Development.
Not surprisingly, one of these viral images was of dubious veracity.1 But the other, the Instagram ad, is quite authentic, and exemplary of the hard push by business schools, continuing education divisions, professional programs, and endowed academic centers to enroll non-traditional students in “business science”2 courses and certifications (many of them using “AI” as a buzzword). These programs are unaccredited, not financial aid eligible, primarily online, asynchronous, partially automated, and often delivered entirely by the university’s private partners.
There are several EdTech companies who are aggressively seeking such partnerships, none more successfully than Ziplines Education, who is the contractor behind the Wisconsin certificate. Ziplines boasts of more than thirty “university partners” on its website, and its “AI Prompting Certificate Course,” launched less than a year ago, is now available through over a dozen institutions.
The tragicomic theme of this weekend’s social media discourse - encapsulated by Tressie McMillan Cottom’s skeet above - is that “prompt engineering” and “AI prompting” are just euphemisms for writing, though a specifically antisocial, transactional mode of writing.
ChatGPT and other “AI” applications marketed aggressively to university students have been the bane of writing instruction since 2023, with certain Silicon Valley thought leaders and EdTech carnival barkers forecasting the obsolescence of academic writing pedagogy and even textual literacy altogether. It would be characteristic of the digital disruption era if - following the model of ridesharing and homestay apps - the long-term impact of LLM chatbots is not to revolutionize a service sector, but merely to deskill and precaritize the existing workforce of educators and impose technofeudal rents on students and schools.
It may very well be that “prompt engineering” will eventually be legitimized enough to be proposed as an automated, enshittified alternative to academic writing (a ubiquitous general education requirement which neoliberal university administrators have long fantasized about sunsetting). These certificate programs are no doubt imagined by some as a step in that direction.
But I see the more imminent catastrophe as the rebirth of for-profit colleges, tapeworm-like, inside public universities. To cite another Cottom truism: “EdTech is a Trojan horse for elite capture of public resources. Every time.”
The adoption of what Cottom deems “Lower Ed” to HigherEd - that is, largely prefabricated and automated education products marketed to non-traditional and usually marginalized students, and delivered by underemployed gig workers - has long been foreseen by Cottom and Dominique Baker, but many of the attempts since the collapse of for-profit colleges a decade ago have been so transparently shoddy they were unable to gain traction with consumers, or were dead on arrival due to labor organizing, student protest, and/or tighter state and federal regulations.
Ziplines Education, contrastingly, is expanding rapidly with zero friction in the anti-education, grifter-friendly environment of 2025.
But Ziplines is not a new enterprise. It is a “startup” only by the perverse definition of such things. It was founded more than a decade ago. It has already gone through two rebrandings and three rounds of venture fundraising. The “micro-university” which would eventually become Ziplines was hatched and germinated inside a venture capital firm.
This Trojan horse has been a long time in the making.
BendPoly: The 9th Semester of Business Science in the Fourth Industrial Revolution
When I say Ziplines was born inside a venture capital firm, I don’t mean that’s who funded it, or seeded it, or advised it, although all those things are true as well. I mean, it was conceived, built, and launched in 2015 by one of Wildcat Ventures’ founding partner, Bruce Cleveland, with the express intention of “spearhead[ing] the first successful pilot of a new micro education model.”
It was initially branded as BendPoly (short for Bend Polytechnic Academy). It had a small physical plant3 in Bend, Oregon, where it delivered an in-person curriculum to a handful of locals, most of them college graduates with liberal arts degrees, who were looking to make themselves more appealing in Oregon’s expanding tech labor market.
Cleveland is a grand narrator and jargoneer of the type who have been essential to late capitalist techno-utopianism and the cultural lionization of private fund managers since 2007. His visionboard pitch for BendPoly makes it sound like the founding of the Starfleet Academy. It would offer a “9th semester” immersion to prospective “business scientists” who would earn a “microdegree” showing their preparedness for the “fourth industrial revolution” which was “already happening”!!!
In truth BendPoly trained a few dozen already well-educated young professionals to use the software applications then preferred by regional employers. Small groups of students met “synchronously” in a classroom space with an instructor who gave them hands-on demonstrations, answered questions, assessed their progress, and even helped with post-certification placement and career counseling. To uses the jargon which would be popularized in its wake, BendPoly was devoted to closing the “skills gap” in a specific, geographically-defined labor market.
This program was successful at its small scale and highly replicable. Many community colleges and regional universities already had analogous offerings, sometimes subsidized by local employers. BendPoly eventually rented space from the nearby private religious college who saw the merit in this humble enterprise, an early iteration of what would become the core strategy: renting association with traditional collegiate brands.
But BendPoly had two serious shortcomings as a venture. First, though there was substantive demand - what Cleveland prefers to call “Minimum Viable Repeatability” - the margins were thin. Wildcat Ventures “discovered” that durable truth of for-profit education: the costs of labor, space, and technology press up against the diminishing returns consumers experience when class sizes grow. Scaling such a business on the model expected by VCs is exceedingly hard.
Wildcat Ventures was also sensitive to the intense consumer and regulatory hostility facing for-profit education in the waning years of the Obama administration. If they sold BendPoly as an alternative, competitor, or corrective to the traditional not-for-profit college experience, they would draw scrutiny from advocacy groups, media, and government, even if they made no effort, as previous for-profits had, to establish equivalent accreditation.
Of course, simply continuing to run a modest, but sustainable vocational program for their local community wasn’t an option they seriously considered.
GreenFig & Rafter: Across The Traction Gap to The Powered By Model
At the beginning of 2017, Wildcat Ventures rebranded BendPoly, which it started characterizing as a successful incubator or Beta Test of the microdegree’s market viability. Wildcat made its first official seed investment in what they now called GreenFig. Cleveland, given the title of co-founder, described the new business as follows:
Greenfig University is a micro-university offering microdegrees in applied business science and what we want to do is establish strategic partnerships with the market leading providers.
The first of these “strategic partners” was Marketo, an industry leader in marketing software. It just so happens that Cleveland, when he was a partner at another VC firm, InterWest, had been Marketo’s first investor and had ushered the company to an IPO which gave it $724 million in market capitalization.4
GreenFig never saw corporation-embedded micro-universities (something which already existed, for instance, at Goldman Sachs) as a durable business model. But, by using GreenFig to power their in-house training operations, Marketo was doing their angel investor a favor, legitimizing his new venture and helping it surmount what Cleveland calls the “traction gap,” the treacherous period for startups between the introduction of their product and their capacity to scale.
What signaled GreenFig’s actual strategy was not the much-publicized partnership with Marketo, but the simultaneous recruitment of Sara Leoni to be the company’s CEO and public face. In time, as Cleveland left Wildcat to launch his own Traction Gap fund, and Wildcat’s seed investment gave way to new rounds of venture funding, GreenFig’s narrative would come to center entirely on Leoni. By 2020, when GreenFig was part of a class of “accelerators” funded by Village Capital, Bruce Cleveland, Wildcat Ventures, and BendPoly had been exorcized from the story the company told about itself.5
GreenFig was promoted as a “woman founded” company, with Leoni spinning a lean-in-feminist origin story about how her kids, despite graduating from highly-ranked business schools, lacked the technical skills necessary for the job placements she wanted for them. She took the initiative to build a company that would fill the skills gap left by traditional education in the digital economy which was in a state of perpetual disruption by new technology.
But from the perspective of the venture capital firm that originally hired her, what Leoni had was a proven track record convincing universities to willingly bring EdTech Trojan horses inside their walls.
Leoni came to Wildcat’s attention as the CEO of Rafter (formerly BookRenter), a textbook rental service that had been a darling of EdTech investors between 2009 and 2014.6 Rafter was in the process of being scrapped and sold for parts to their leading competitor, eCampus, when GreenFig was launched. The amount of that sale has never been disclosed, but even assuming investors took a heavy loss, the sales model Rafter innovated was valuable enough that eCampus imitated it (as eCampus All Access), then expanded that imitation after they acquired Rafter’s assets. Leoni was thus an established disruptor.
Leoni called the Rafter sales model “Textbooks-in-Tuition.” At the institutions with whom Rafter contracted, textbook costs would be built into tuition and the textbook packages designed by instructors would be automatically delivered to students by Rafter as soon as they registered for the course. Colleges got assurance that every student would have required course materials from day one at a discount rate. Rafter got a captured consumer base, revenue certainty, and the likelihood that their costs would decrease and profits increase as their inventory of textbooks and licenses grew.
Crucially, students at Rafter-contracted institutions were more likely to be able to use student loans to cover their textbook costs. As such, Rafter was also a middling Ponzi austerity scheme, as it offloaded risk to schools, who had to decide whether and how much to hike tuition in order to pay Rafter, or had to find savings elsewhere in their budgets. In many cases, the schools became fiscal passthroughs between student lenders and Rafter, while also providing free customer service.7
Rafter ended up in the overflowing dustbin of EdTech startups circa 2016, but Wildcat Ventures must have been impressed by the fact that Leoni had, regardless of Rafter’s final fate, masterminded a Ponzi austerity scheme and successfully negotiated a series of what looked like highly advantageous strategic partnerships with colleges and universities. At GreenFig, her job would be to create a Textbooks-in-Tuition-style sales model for business science microdegrees. This she did, branding it the “powered by model.”
“Powered by” is an allusion to the fine print which appears on some promotional and admission materials associated with courses and certifications which are offered through accredited institutions, but produced and distributed by private companies. This fine print does not necessarily say powered by, but offers some similarly tepid acknowledgment that the educational institution may not fully control the content or delivery of the course. The landing page for “AI Prompting” at Wisconsin, for instance, says “in partnership with Ziplines.”
In my opinion, a more accurate way to describe the “powered by model” would be brand equity rental. GreenFig was paying its university partners, usually as a percentage of the tuition fees they generated, to borrow their names, logos, color schemes, domains, social accounts, etc. Once students enrolled through the branded admissions portals, they became part of an integrated pool of students from across the country who were all taking the same online course conceived and administered by GreenFig.
The “powered by model” is a truly absurdist role reversal. A private, unaccredited company founded and run by sales and marketing professionals is responsible for the (pseudo)educational coursework, while the accredited university is employed only for its sales and marketing functions, getting paid by commission on the headcount of students who enroll from their branded portal. University partners are incentivized to flex their brand power and use their proprietary data, advertising budgets, and sales forces to maximize this commission, while Ziplines provides cookie-cutter landing pages and highly reproducible microdegrees, the content of which is largely created by gigworkers.8
It’s no wonder that in his 2020 profile of Leoni for Forbes, David Newton predicted that the “powered by model” which GreenFig was then rolling out would be “a game-changer in the education business” because
There may a good deal of runway in giving colleges content and curricula they cannot or do not want to build while using their branding, community roots, infrastructure and students.
GreenFig was no longer a for-profit institution selling an “alternative to college,” they were “part of it,” having signed contracts with ten institutions as of the Forbes profile’s publication. GreenFig was using many of the things which gave college value to create something which colleges, rightly, did not want. The traction gap had been crossed. The horse was inside the walls.
Ziplines Education: The New New Thing That Might Save Middle America
Of the GreenFig university partners mentioned in the 2020 Forbes profile, only one, Grand Valley State University, in Michigan, retains its contract with Ziplines. But since its rebrand in February of last year, the company has enjoyed a period of unprecedented growth, signing at least eleven new partnership agreements, including with enormous, highly-regarded public flagships in Texas, Utah, and Wisconsin. The rebrand also immediately preceded the company’s largest capital infusion, a $6.4 million investment led by Greg Gretsch of Jackson Square Ventures, the managing director behind, among other things, the IPO of Upwork and the sale of grocery delivery startup, Cornershop, to Uber.
The size of this investment provoked Newton to write a follow-up, self-congratulatory article for Forbes, in which he noted, accurately
[$6.4 Million is] an impressive number for most any Series A, but for education market investing, which has been suppressed over the past few years, $6.4 million is a spotlight figure. The investment, led by Jackson Square Ventures in California, is an endorsement for both founder and foundational idea.
The for-profit college bubble burst in 2015, and EdTech investing dropped off a cliff after 2021, but Ziplines’ recent success augers well for the reemergence of both, as does, of course, the anti-HigherEd agitprop, ideological defunding, and deregulatory free-for-all of 2025.
Behind the Ziplines rebrand was a refining of the company’s marketing strategy to thrive in this environment. It has three tenets.
Appeal to the economic anxiety created by AI hype machine.
Target prospective students in geographic regions far removed from established tech centers.
Rent the brand equity of public universities associated with regional pride, while simultaneously sowing doubt in the their legitimacy.
Ziplines appeal to economic anxiety is just an updating of the narrative which the company has been forwarding since its BendPoly days: professions change fast because of rapidly-evolving technology, your education and skills are in a state of perpetual degradation, stay frosty by learning the latest trends in business science (a.k.a. how to use planned obsolescent software).
This narrative is even more compelling to workers who are being told everyday by celebrity entrepreneurs, business journalists, and sometimes their own employers, that entire professions are going to disappear on time horizons of a few years or even months. Leoni offers the AI-infused version of her company’s durable sales pitch in a Q&A with Gretsch for the Jackson Square website:
The reality is that with the acceleration of technology, platforms, and innovations, such as generative AI, embracing the rate of change is table stakes. Jobs are changing. The skills required are evolving. What we know today will be obsolete a year or two from now. Our goal is to continue to innovate and help the workforce stay relevant and in demand — and confident about taking on what’s ahead.
Even in her rose-colored explanation of the new branding, there is a subtext of accelerationism, labor precarity, and economic anxiety:
Why Ziplines? Learning is a journey — it’s intimidating, exhilarating, challenging, and continuous. When you’re ziplining, you have a starting point and an endpoint. You soar from tree to tree with a distinct goal in mind. You’re focused on accomplishing something efficiently in swift, impactful, measurable timeframes. Where one zipline ends, another begins. Just like continuous learning.
But the more concrete change associated with the Ziplines rebrand is who the company is marketing this vision to, and therefore which kinds of university partners it prefers.
GreenFig’s university partners, in keeping with the BendPoly model, were mostly from places that already had thriving tech sectors (UC-Santa Cruz, for instance). But in the interviews that followed from the rebrand and Series A fundraising, Leoni repeatedly alludes to “Middle America” and to Ziplines’ most desirable partners being community colleges and state universities, not “Ivy Leagues.” In other words, Ziplines doesn’t want to rent the brands of the kinds of institutions being most aggressively smeared and targeted for defunding.
But also, a durable truth of the “powered by model,” Ziplines recognizes that despite the enduring, well-funded anti-education campaigns of the past decade, colleges and universities are still far more trusted by the general public than tech companies and VC firms. And that approval rises with proximity. Even people who are willing to accept that famous, centuries-old campuses of New England and the Mid-Atlantic are now havens of communist indoctrination don’t think the same is true of the nearby campuses where they send their kids, and whose football team they root for on Saturdays.
Ziplines is renting the brand equity of regionalized public higher education in the aggregate.
86% of Ziplines advertised university partners as of this writing are public institutions. 69% are in the South, Southwest, or Midwest regions, including 82% of those signed since the rebrand from GreenFig.
EdTech is not only always a Trojan horse for elite capture of public resources; it is also always a project in delegitimizing the project of public education itself. Even as Ziplines recognizes and rents the brand equity of schools like Wisconsin and Temple, it preaches their inadequacy and obsolescence to investors, and purposefully erodes the public’s trust in them when it disguises prefab, generic courses behind local, bespoke brands.
The “powered by model” has been embraced precisely because public universities are being slowly starved of public funding, and their desperation for diversified revenue streams creates opportunities for private capture and Ponzi austerity.
The evolution of education which companies like Ziplines imagine is not one of mutual flourishing with their university partners, but one in which those partners can be retired and replaced by automated, scalable, unaccredited online education.9
As I have written before, the EdTech Griftopia prophesized by venture capital and private equity is one in which “‘lifelong learners’ pay for the opportunity to donate their labor to businesses who rebrand entry-level drudgery as cutting-edge training” or in which students are “nightmarish figures trapped in a never-ending Udemy slideshow…that becomes obsolete before it can be completed. Courses in anything, from computer programming to communications to cosplay, are created by anybody, from televangelists to social media influencers to PR shills.”
A Ziplines Education is one in which your life is a series of moderately-priced microdegrees offered by a for-profit college from which you can graduate every few months, but you can never leave.
The First Squawk fintech influencer account on X alleged Huang had “recently told CNBC” when, in fact, their weak paraphrase was of Huang’s comments during a podcast interview with Vox video editor, Cleo Abrams, from January. Huang never uses the phrases “AI literacy” or “prompt engineering,” nor does he suggest that this kind of “top skill” should be formally integrated into any curriculum. And though he does mention students, he also emphasizes that “learning AI” is something “everybody” should be concerned with. What he says is,
“I were a student today the first thing I would do is to learn AI. How do I learn to interact with ChatGPT, how do I learn to interact with Gemini Pro, and how do I learn to interact with Grok? Learning how to interact with with AI is not unlike being someone who is really good at asking questions. You're incredibly good at asking questions and prompting AI is very very similar. You can't just randomly ask a bunch of questions and so asking an AI to be assistant to you requires some expertise and artistry and how to prompt it. And so if I were a student today, irrespective whether it's for for math or for science or chemistry or biology or doesn't matter what field of science I'm going to go into or what profession, I'm going to ask myself, how can I use AI to do my job better? If I want to be a lawyer, how can I use AI to be a better lawyer? If I want to be a better do doctor, how can I use AI to be a better doctor? If I want to be a chemist, how do I use AI to be a better chemist? If I want to be a biologist, I how do I use AI to be a better biologist? That question should be persistent across everybody. And just as my generation grew up as the first generation that has to ask ourselves, how can we use computers to do our jobs better?…The next generation doesn't have to ask that question but it has to ask the obviously next question, how can I use AI to do my job better? That is start and finish I think for everybody. It's a really exciting and scary and therefore worthwhile question I think for everyone.”
Business science, as a euphemism for proficiency with commercial software applications, is an invention of EdTech. Its origins date back to the Web 1.0 era, but last month (June 2025), tellingly, it peaked as a search term, according to Google Trends.
I have not been able to confirm this, but I believe BendPoly’s initial courses took place inside a technophile coworking space which Cleveland also owned.
Marketo was thereafter taken private again in a $1.8 Billion deal by Robert Smith’s Vista Equity, which sold it one year later to Adobe for $4.8 Billion.
I did not make any effort to uncover the story behind Cleveland’s departure from Wildcat or his stepping away from the company he founded, but it seems worth noting that on the Traction Gap Partners website, he still lists himself as a former “founder and CEO of Ziplines Education.”
According to the Failory “startup cemetary,” Rafter raised $86 million before it went offline in 2016 and was soon after sold to eCampus.
Since the schools selected Rafter as their exclusive textbook provider and their professors were choosing the rental packages, it was the schools that got blamed when Rafter didn’t deliver materials up to students’ expectations.
The advertised “course creator” for the AI Prompting course is Danoosh Kapadia, a freelance “AI Educator and Growth Strategist” from San Francisco (according to his LinkedIn profile).
This fantasy abounds in the Q&A between Leoni and Gretsch.
Hi Matt, I haven't yet read this post in its entirety, mostly because it's so thought-provoking that it's taking me some time to process. One of the thoughts it's provoked so far, though, is your argument's resonance with a piece by a long gone Columbia sociologist, Robert Lynd, on the Truman Commission's Report, Higher Education for American Democracy. Lynd's 1948 essay, "Who Calls the Tune?," appeared in the Journal of Higher Education (19.4). (I'm happy to send a pdf, if you can't access that journal.) Lynd's perspicacity astounds: as the Report (famously) called for higher ed's help in sustaining democracy, Lynd states plainly that our democracy--this is immediate post-WWII--was already, and perhaps fatally, hampered by capitalism. How could higher ed support democracy, then, in its post-war form? What higher ed needed to do was to question the capitalistic incursion and rethink what democracy was/could be otherwise.
I've engaged Lynd's critique in my research into higher ed's role in the implementation of Am foreign policy in the early Cold War. As a lit prof, I am curious about how this role shaped the place of the humanities in higher ed. And, being at MSU, I am particularly concerned with the land-grants' participation. C. Cilano
Important info that helps us to understand venture capital's aims and damage already done to education. I think it's useful to see the cleavage between those supporting Trump's full privatization of education with tech and the new neoliberal project to privatize, hollow out in a "public" system. AFT's deal with OpenAI and Anthropic to help teachers use AI straddles both.