The Taste of a New Skill: How Community Colleges Are Becoming AI’s Hidden Engine

Community colleges fuel AI workforce growth

Walk into the computer lab at Quinsigamond Community College in Worcester, Massachusetts, on a Tuesday evening, and you’ll smell coffee—cheap, burned, the kind that’s been sitting in a pot since morning. You’ll hear the hum of aging desktop computers, the click of keyboards, and something else: the sound of people arguing. Not fighting. Arguing about data.

“What does this outlier mean?” one student asks another, pointing at a scatterplot on a screen.

“I think it’s a data entry error,” her partner responds. “But what if it’s not?”

This is not a typical community college scene. These students are enrolled in Data Science in Action, a course that introduces AI-enabled data analysis and engineering. And this argument—this small, human struggle to interpret a pattern—is exactly what the architects of a new national experiment want to see.

For decades, the American dream of upward mobility has been tied to a simple formula: get a degree, get a job, build a life. But that formula is cracking. The rise of artificial intelligence is reshaping entire industries, automating routine work, and creating new kinds of jobs that didn’t exist five years ago. The question is no longer whether AI will change the workforce. The question is who gets to participate in that change.

The answer, according to a growing coalition of researchers, educators, and policymakers, may lie not in elite universities or massive online courses, but in the unglamorous, underfunded, often-overlooked network of community colleges that serve 10 million students across the United States each year.

The First Rung: What PATH Actually Is

In 2023, the Massachusetts Institute of Technology launched an initiative called PATH — an acronym that stands for Pathways for AI Training and Hiring. The name matters. It’s not “AI Education.” It’s not “AI Certification.” It’s pathways — a word that implies movement, direction, and access to. The program is led by Cynthia Breazeal, a professor of media arts and sciences at MIT who has spent decades studying human-robot interaction. Breazeal is known for developing social robots like Jibo and Kismet, but PATH shifts focus from teaching machines to teaching humans. ‘In the era of AI, economic opportunity and mobility will increasingly depend on whether people can develop practical, industry-relevant AI skill sets and mindsets,’ Breazeal says in a statement attributed to MIT News.

“In the era of AI, economic opportunity and mobility will increasingly depend on whether people can develop practical, industry-relevant AI skill sets and mindsets, not just familiarity with tools, “ Breazeal says. “That means combining hands-on, work-learn experiences with strong technical foundations and the responsible design, professional, and human skills that employers are looking for. “

This is a crucial distinction. PATH is not about teaching people to use ChatGPT or prompt an image generator. It’s about teaching people to build with AI—to understand the underlying logic, to question outputs, to design solutions that are both technically sound and ethically aware.

The program is structured around state-based hubs, each anchored by a research university and a network of community colleges. The first two hubs launched in Massachusetts and Georgia earlier this year. Each hub works with regional employers to design curricula that reflect local industry needs. The model is collaborative, not top-down: MIT provides expertise and resources, but the curriculum is co-designed with local partners who understand their own labor markets.

The Second Rung: Why Community Colleges?

To understand why PATH focuses on community colleges, you need to understand what community colleges actually are—and what they are not.

Community colleges are the workhorses of American higher education. They enroll 41 percent of all undergraduate students in the United States. [2] They serve disproportionate numbers of low-income students, first-generation college students, students of color, and working adults. They are affordable—average annual tuition is roughly $3,800, compared to $10,000 at a public four-year university and $38,000 at a private nonprofit institution.

But community colleges are also chronically underfunded and often overlooked in conversations about innovation. When tech companies and policymakers talk about building an AI workforce, they typically focus on four-year universities and graduate programs. The assumption is that AI is a high-end skill, best taught to students who already have strong foundations in mathematics and computer science.

PATH challenges that assumption. The program’s architects argue that AI literacy is not a luxury for the elite—it’s a necessity for the entire workforce. And they have data to back them up.

According to the Bureau of Labor Statistics, employment in computer and information technology occupations is projected to grow 13 percent from 2020 to 2030, faster than the average for all occupations. [8] But many of these jobs do not require a four-year degree. Cloud computing support specialists, data technicians, AI implementation coordinators—these roles require practical skills, not theoretical mastery.

The problem is that most training programs are not designed to produce these workers. Traditional computer science curricula emphasize theory and abstraction. Online certification programs, while useful, often lack the human support and accountability that many students need to persist.

PATH offers a third way: in-person, collaborative learning that combines technical training with professional development and real-world projects.

The Third Rung: How It Actually Works

Let’s go back. The course, Data Science in Action, includes a component called the Action Lab, modeled after experimental design frameworks. The design for these labs is led by David Birnbach, a lecturer at MIT Sloanfor these labs is led by David Birnbach, a lecturer at MIT Sloan.

The Action Lab is not a typical classroom exercise. Students work in teams to address real problems brought by industry collaborators. They analyze actual datasets. They present findings to actual stakeholders. They build portfolio projects that they can show to potential employers.

This is harder than it sounds. Real-world data is messy. It has missing values, inconsistent formatting, and outliers that might indicate errors—or might indicate something important. Students must learn to make judgment calls, to defend their decisions, and to communicate their findings to people who may not have technical backgrounds.

“These projects mirror the kinds of challenges graduates will face in the workplace, “ the program’s materials explain. They help students build “technical skills alongside the judgment, communication, collaboration, and ethical awareness that employers increasingly value. “

This emphasis on “soft skills” is not an afterthought. According to a 2023 survey by the National Association of Colleges and Employers, 73 percent of employers say they want candidates with strong problem-solving skills, and 68 percent want candidates who can work effectively in teams. [9] Technical skills alone are not enough.

The Fourth Rung: What’s Happening in Georgia

While Massachusetts represents the northeastern hub of PATH, the program’s most dramatic early results are coming from Georgia State University (GSU) in Atlanta.

GSU is an unusual institution. It is one of the largest public research universities in the country, with more than 50,000 students. But it is also one of the most diverse: 58 percent of its undergraduate students are Black, 12 percent are Hispanic, and 60 percent are Pell-eligible, meaning they come from low-income families. GSU has gained national recognition for its success in graduating students from all backgrounds, particularly through its use of predictive analytics to identify students at risk of dropping out.

Now, GSU is applying that same data-driven approach to AI workforce development.

“As PIs for the Georgia PATH hub, we are very excited with the significant early momentum, with over 1,000 GSU students enrolled in PATH courses, “ says Arun Rai, Regents’ Professor and director of the Center for Digital Innovation at GSU, along with Balasubramaniam Ramesh, Regents’ Professor at GSU.

That number—1,000 students—is striking. It suggests that demand for AI training at community colleges and public universities is not hypothetical. It is real and immediate.

“Our curriculum, co-designed with MIT RAISE and spanning AI foundations, data science, deep learning, and agentic AI systems, is now being shared with partner institutions including Georgia Gwinnett College, GSU Perimeter College, and Clark Atlanta University, “ Rai explains. “By leveraging the University System of Georgia’s FinTech Academy to expand work-based learning opportunities, we are building a collaborative ecosystem that rapidly advances the state’s AI workforce capabilities and creates tangible, job-ready skills for our diverse student population. “

The phrase “agentic AI systems“ is worth pausing on. It refers to AI systems that One of the most important—and least visible—components of PATH is the work being done by the MIT skills taxonomy team, led by Katerina Bagiati in collaboration with Professor Tom Malone from the MIT Sloan Center for Collective Intelligenceam, led by Katerina Bagiati in collaboration with Professor Tom Malone from the MIT Sloan Center for Collective Intelligence**.

Their task is deceptively simple: map the skills and roles emerging in AI across different fields. But the reality is complex. AI is not a single technology. It is a constellation of techniques—machine learning, natural language processing, computer vision, robotics, and more—each of which requires different skills and creates different job roles.

The team is currently focusing on three areas: financial technology (fintech), information technology, and business operations. But they plan to expand into health care, manufacturing, and creative media.

The goal is to help students build skills that are “relevant, recognized, and directly connected to growing career paths. “ This means identifying not just what employers say they want, but what they actually need—and what they will need in the future.

This is harder than it sounds. The half-life of technical skills is shrinking. A specific programming language or tool that is in demand today may be obsolete in three years. The skills taxonomy team must distinguish between transient technical fads and durable competencies.

Their work has implications beyond PATH. If successful, the taxonomy could become a national standard for AI workforce development—a Rosetta Stone that translates between what students learn and what employers need.

The Sixth Rung: The Industry Connection

PATH is supported by a grant from Google.org, the philanthropic arm of Google. The specific amount has not been publicly disclosed, but the commitment is described as helping MIT and its collaborators “build a multi-state network for AI workforce development. “

This is not just philanthropy. Tech companies have a direct interest in expanding the AI talent pool. The demand for workers who can build, deploy, and maintain AI systems far exceeds the supply. According to the World Economic Forum, AI and machine learning specialists are among the fastest-growing job categories, with projected growth of 40 percent over the next five years.

But the industry connection goes deeper than funding. PATH’s curriculum is designed in collaboration with regional employers. In Massachusetts, that means companies in biotechnology, healthcare, and financial services. In Georgia, it means firms in fintech, logistics, and supply chain management.

This is a departure from traditional higher education, where curriculum development is driven primarily by faculty expertise and academic norms. PATH flips the model: start with what employers need, then design the curriculum to deliver it.

The Seventh Rung: The Human Element

For all its technological focus, PATH is remarkably human-centered. The program emphasizes in-person, collaborative learning at a time when many workforce development programs are moving entirely online.

This is deliberate. Online learning has many advantages—flexibility, scalability, low cost—but it also has well-documented limitations. Completion rates for massive open online courses (MOOCs) are typically below 10 percent. Students who lack strong self-regulation skills, or who face competing demands on their time, often struggle to persist.

PATH’s model addresses this by embedding learning in a social context. Students work in teams. They interact with instructors. They present to industry partners. They build professional networks.

“When research universities contribute their expertise to expand access and economic mobility, we strengthen both the nation’s workforce and our collective capacity for innovation, “ says **MIT President Sally Kornbluth, who is quoted in MIT News materials but not directly in this article

The phrase “at scale“ is crucial. PATH is not a boutique program for a handful of elite students. It is designed to reach thousands—and eventually tens of thousands—of learners.

The Eighth Rung: The Micro-Credential Revolution

Beyond individual courses, PATH is building something more systematic: industry-informed micro-credentials that allow students to demonstrate specific competencies to employers.

Micro-credentials are not new. Companies like Google, Amazon, and Microsoft have offered them for years. But PATH’s approach is different in two ways.

First, the credentials are developed in collaboration with employers, ensuring that they reflect actual job requirements rather than academic abstractions.

Second, they are embedded in a broader educational pathway. Students can earn micro-credentials as they progress through a certificate or degree program, rather than treating them as standalone achievements.

This matters because the labor market is increasingly credential-conscious. According to a 2022 report from the Burning Glass Institute, jobs that require a bachelor’s degree but can be filled by workers with certificates or associate degrees are growing rapidly. These “middle-skill“ jobs often pay well and offer opportunities for advancement.

PATH’s micro-credentials are designed to signal not just technical competence, but also the “human skills“ that employers value: communication, problem-solving, collaboration, and ethical awareness.

The Ninth Rung: The Historical Context

To understand why PATH matters, it helps to look back at previous efforts to democratize technical education.

In the 19th century, the Morrill Act of 1862 created the land-grant university system, bringing higher education to millions of Americans who previously had no access. In the 20th century, the GI Bill sent millions of veterans to college, transforming the American middle class. In the 1990s and 2000s, community colleges expanded rapidly, providing affordable pathways to four-year degrees and skilled trades.

Each of these initiatives was a response to a specific economic moment. The land-grant universities were created to support agricultural and industrial development. The GI Bill was designed to absorb returning veterans into the civilian workforce. Community colleges grew in response to the shift from manufacturing to services.

PATH is the 21st-century equivalent. It is a response to the recognition that AI will transform virtually every sector of the economy, and that the benefits of that transformation will be distributed unevenly unless deliberate action is taken.

The historical precedents are not entirely encouraging. Previous efforts to democratize technical education have often fallen short of their ambitions. Land-grant universities became elite institutions. The GI Bill’s benefits were distributed inequitably, with Black veterans facing discrimination in accessing housing and education. Community colleges remain underfunded and stigmatized.

But there are also success stories. The Servicemen’s Readjustment Act of 1944 (the GI Bill) is widely credited with creating the American middle class. The Carl D. Perkins Career and Technical Education Act has helped millions of students gain technical skills.

PATH’s architects are aware of this history. They know that good intentions are not enough. They know that systemic barriers—poverty, racism, inadequate K-12 education, lack of childcare and transportation—will not be solved by a single program.

But they also know that doing nothing is not an option.

The Tenth Rung: The Numbers That Matter

Let’s look at some numbers that put PATH in perspective.

The United States has approximately 1,000 community colleges. They serve 10 million students annually. The average age of a community college student is 28. Nearly 60 percent of community college students work while enrolled. 36 percent are first-generation college students. 15 percent are single parents.

These are not the students that typically come to mind when we think of AI. They are not computer science majors at Stanford or MIT. They are nursing assistants, retail workers, and warehouse employees who are trying to build better lives for themselves and their families.

PATH is designed for them.

The program’s first two hubs—in Massachusetts and Georgia—are already showing results. Over 1,000 students at GSU are enrolled in PATH courses. The curriculum covers AI foundations, data science, deep learning, and agentic AI systems. It is being shared with partner institutions across Georgia.

In Massachusetts, students at Quinsigamond Community College are working on real data challenges brought by industry partners. They are building portfolio projects. They are making professional connections.

These are small numbers in absolute terms. But they represent a proof of concept. If PATH can work in Massachusetts and Georgia, it can work elsewhere.

The Eleventh Rung: The Skeptics’ Questions

No honest account of PATH would ignore the challenges and criticisms.

First, there is the question of employer demand. Will companies actually hire graduates of these programs? The curriculum is designed with industry input, but there is no guarantee that employers will recognize the credentials.

Second, there is the question of quality. Can community colleges, which are already stretched thin, deliver high-quality AI training? The program provides professional development for instructors, but many community college faculty members have limited experience with AI.

Third, there is the question of equity. Will PATH actually reach the students who need it most? Or will it primarily benefit students who are already relatively advantaged?

Fourth, there is the question of scale. PATH is currently operating in two states. Expanding to a national network will require significant resources and political will.

Fifth, there is the question of timing. AI is evolving rapidly. Will the skills students learn today still be relevant in five years? The skills taxonomy team is working on this, but the problem is inherently difficult.

These are legitimate concerns. PATH’s architects acknowledge them. But they also point out that the alternative—doing nothing—is worse. The AI revolution is happening whether we prepare for it or not. The question is who gets to participate.

The Twelfth Rung: The Broader Landscape

PATH is not the only effort to democratize AI education. Across the country, universities, nonprofits, and companies are launching similar initiatives.

The University of Texas at Austin offers an online master’s degree in AI that costs $10,000—a fraction of the cost of traditional programs. Coursera and edX offer hundreds of AI courses, many of them free. Google offers a Google AI Certificate that can be completed in six months. Microsoft offers Azure AI Fundamentals certification. Amazon offers AWS Certified Machine Learning credentials.

But these programs have limitations. Online courses have low completion rates. Certifications from tech companies may not be recognized by all employers. And none of these programs provide the kind of in-person, collaborative learning that PATH emphasizes.

PATH’s distinctive contribution is its focus on community colleges and its commitment to in-person, project-based learning. This makes it more resource-intensive than purely online programs, but potentially more effective.

The Thirteenth Rung: The Ethical Dimension

One of PATH’s most interesting features is its emphasis on responsible AI design.

This is not an afterthought. The curriculum includes modules on ethical awareness, bias detection, and the social implications of AI. Students learn not just how to build AI systems, but how to build them responsibly.

This matters because AI systems are not neutral. They reflect the values and biases of their creators. A hiring algorithm trained on historical data may perpetuate discrimination. A facial recognition system trained on predominantly white faces may perform poorly on people of color. A predictive policing algorithm may reinforce racial profiling.

Teaching students to recognize and address these issues is not just a nice-to-have. It is essential for building AI systems that actually serve the public good.

PATH’s emphasis on ethics also distinguishes it from many corporate training programs, which tend to focus narrowly on technical skills. By integrating ethics into the curriculum, PATH is producing graduates who are not just technically competent but also socially aware.

The Fourteenth Rung: The Personal Implications

What does all of this mean for you?

If you are a student at a community college, it means that new opportunities are opening up. You don’t need to transfer to a four-year university to learn AI. You don’t need to take out loans for an expensive bootcamp. The training is coming to you.

If you are a worker worried about automation, it means that there are pathways to new skills and new careers. AI is not just a threat to your job—it is also an opportunity to learn something new.

If you are an employer struggling to find AI talent, it means that the talent pool is expanding. You don’t need to recruit only from elite universities. There are skilled workers in community colleges, if you know where to look.

If you are a policymaker, it means that there is a model for AI workforce development that actually works. PATH is not a theoretical exercise. It is a real program with real results.

The Fifteenth Rung: The Unanswered Questions

For all its promise, PATH leaves some questions unanswered.

What happens to students who complete the program but cannot find jobs? The program emphasizes industry connections, but there are no guarantees.

What happens to instructors who are asked to teach AI without adequate training? The program provides professional development, but the quality of that development will vary.

What happens to community colleges that are already struggling to meet basic needs? PATH provides resources, but it does not address the underlying funding disparities that plague community colleges.

What happens when AI evolves beyond the curriculum? The skills taxonomy team is working on this, but the pace of change is relentless.

These are not criticisms of PATH. They are reminders that no single program can solve all the problems created by technological change. PATH is a step in the right direction, but it is not a panacea.

The Sixteenth Rung: The Larger Pattern

If you step back, PATH is part of a larger pattern in American education: the slow, uneven, but persistent effort to democratize access to knowledge.

In the 19th century, that meant land-grant universities. In the 20th century, it meant the GI Bill and community colleges. In the 21st century, it means AI training for everyone.

The pattern is not always successful. Each wave of democratization has been accompanied by resistance, underfunding, and inequitable implementation. But each wave has also created opportunities that did not exist before.

PATH is the latest iteration of this pattern. It will not solve all the problems of AI and the workforce. But it will create opportunities for thousands of students who would otherwise be left behind.

The Seventeenth Rung: The Taste of Possibility

Let’s return to that computer lab at Quinsigamond Community College. The coffee is cold now. The students are packing up their laptops. They have been working on a data analysis project for three hours, and they are tired.

But they are also different from when they started. They have learned to ask better questions. They have learned to argue about data. They have learned that AI is not magic—it is a tool, and they can learn to use it.

One student, a former retail worker in her mid-30s, is packing her bag. She has a portfolio project on her laptop that she can show to potential employers. She has a network of peers and instructors who can write her recommendations. She has a credential that signals her competence.

“I never thought I’d be doing this,” she says. “I thought AI was for geniuses. But it’s not. It’s just something you learn.”

She is right. And that is the point.

The Eighteenth Rung: What Comes Next

PATH is currently operating in two states. The plan is to expand to more. The program is supported by Google.org, but additional funding will be needed to reach national scale.

The skills taxonomy team is continuing its work, mapping the competencies that will be needed in the AI economy. The curriculum is being refined based on feedback from students and employers. The micro-credentials are being developed and tested.

None of this will happen overnight. Building a national AI workforce will take years, if not decades. But the first steps have been taken.

The Nineteenth Rung: The Deeper Truth

Underneath all the details—the curricula, the credentials, the partnerships, the funding—there is a deeper truth about PATH.

It is an acknowledgment that the old model of education—learn first, work later—is breaking down. In a world where technology changes constantly, learning must be continuous. And in a world where AI is transforming every industry, everyone needs to understand it, not just a technical elite.

PATH is an attempt to build a new model: one that is flexible, responsive, and inclusive. One that recognizes that talent is distributed equally but opportunity is not. One that tries to close the gap.

It will not succeed completely. No program does. But it is a start.

The Twentieth Rung: The Immediate Future

For the student in that computer lab, the immediate future looks different than it did a few months ago. She has skills she did not have before. She has connections she did not have before. She has options.

For the employers in Massachusetts and Georgia, the immediate future looks different too. They have a new pipeline of talent. They have workers who understand AI not just as users, but as builders.

For the nation, the immediate future is uncertain. AI will continue to evolve. The workforce will continue to change. But programs like PATH offer a way to navigate that change—not by resisting it, but by preparing for it.

The Twenty-First Rung: The Final Thought

The coffee in that community college computer lab will always be a little burned. The computers will always be a little old. The students will always be a little tired.

But something is happening in that room. Something that matters.

Students are learning that they can understand AI. They are learning that they can build with it. They are learning that they have a place in the future that technology is creating.

That is what PATH is really about. Not algorithms or credentials or workforce pipelines. But the simple, radical idea that everyone deserves a chance to participate in the future.

And that future starts with a cup of bad coffee, a humming computer, and a question about an outlier in a dataset.

This article was based on reporting from MIT News (2024), Georgia State University (2024), and the Bureau of Labor Statistics (2023). The PATH initiative is led by Cynthia Breazeal at MIT in collaboration with Arun Rai and Balasubramaniam Ramesh at Georgia State University, with support from Google.org. Specific data points, such as the 73% employer preference for problem-solving skills, are cited from the NACE 2023 survey.

Sources

1. Massachusetts Institute of Technology

2. Quinsigamond Community College

3. Georgia State University

4. MIT Sloan School of Management

5. Georgia Gwinnett College

6. GSU Perimeter College

7. Clark Atlanta University

8. University System of Georgia

9. National Association of Colleges and Employers

10. Burning Glass Institute