‘Vibe coding’ can give us insight into the future of AI

By Jacob Sweet Harvard Staff Writer | Harvard Gazette

It is no longer necessary to know how to code to design a website or app Describe in plain English what the program should do, and an AI agent will do its best to execute the vision—a process called “vibe coding.” The end result may have a lot of limitations, but it will be far superior to what someone without fairly significant technical skills could produce.

Among those exploring new practices are Karen Brennan, Timothy E. Worth is a professor of the practice of learning technology at the Harvard Graduate School of Education, who taught a six-week course on Vibe Coding late last fall. In this edited interview, Brennan details what he tells students and the insights he’s gained about what our future might look like with AI.

What is vibe coding and what was your first experience with it?

Vibe Coding is about creating software with the help of AI – and specifically, creating software where you don’t necessarily understand the code that’s being created. (“Vibe coding” is a term popularized by computer researchers Andrej Karpathi in February 2025.)

Responsibility for understanding the underlying code separates professional software development, which is increasingly AI-assisted, from Vibe coding. As a term, vibe coding can be positive or pejorative, either celebrating the freedom to understand code or underscoring the risks of shirking that responsibility.

My first experience with Vibe Coding was through December 2024 Harvard Initiative for Learning and Teaching-funded research projectI was studying how students were using generative AI in self-directed projects, and a student introduced me to v0 (an AI-powered tool for building web applications and sites).

When I later need to create a website for the same research project, I use v0 to Build it. I was amazed at how quickly I was able to build the site and the quality of what was produced.

You taught a course on Vibe Coding, no prior experience with AI or coding required. What were your hopes for the course and how did it turn out?

My doctoral student Jacob Wolf And I designed one Vibe is a six-week course on coding which we taught late last year, supported by an amazing teaching team. Our hope for the course was to explore this particular sociotechnical moment, where anyone can (in theory) create software in collaboration with AI.

The central question motivating the course was: How do we think about AI as creative partners? We had a different theme each week (create something that tells a story, that makes your life easier, that invites play, etc.) and tried different Vibe coding tools each week (eg, Replit, Figma Make, Clod Code).

The central question motivating the course was: How do we think about AI as creative partners?

The course was clearly not about building professional software; We are focused on experimenting with new creative possibilities.

An important part of course design for us was to create with a critical perspective. Each week we read a classic text from computer science (to remind us that people have been thinking about the opportunities and challenges of AI for decades) and a more contemporary critical piece (to defend against AI hype).

We asked students to put their experiences and lessons in conversation with each other into a final position portfolio. Now that they’ve experienced all this, how do they feel about creating with AI? What do they want to say about it?

I loved the experience — for what our students were able to create and critique, and the opportunity to co-create with the students. 92 students participated and based on course evaluations and our meetings with them, students appreciated both the opportunity to make things for themselves and the critical framework we brought to that work.

It was a huge experiment, and like any first-run course we were figuring out a lot in real time, especially given the ongoing technology changes. But I think the positive feedback we’ve had from students has confirmed that it’s not too early to try to teach it at universities in schools of education.

What are some of the promises and limitations of vibe coding?

The key promise for me is the democratization of creation. Vibe Coding makes software production, the output of code, accessible to more people. You can get an idea and realize that idea without having a degree in computer science or hiring a team of developers. And so it’s changing the economics of experimentation: to understand a thing, you often have to make a thing, and now you can make that thing very quickly. That rapid iteration and tinkering is a way to generate more ideas and unlock creativity.

Being in a school of education, there is also a learning dimension that I find exciting. While Vibe Coding can be a way to avoid CS content knowledge, many of these tools create opportunities for you to inspect and test code implementations. You can peek under the hood!

And you can ask the AI ​​to explain what you’ve put together at whatever level of detail you’re most comfortable with, from more technical explanations to how you can ask it to explain it like you’re a first grader.

Of course, there are many limitations. Environmental impact and cost were two of the concerns we addressed. And as a creator, you are limited by your ability to express your ideas in natural language. Students with CS knowledge or a design background can go further because they can more clearly describe what they expect.

We’ve seen students get stuck in a frustrating loop: prompting the AI ​​for something, the AI ​​producing something that isn’t quite right or something that seems generic, and then students unable to fully articulate the problem and what needs to change. Vibe coding privileges people who are strong verbal communicators, which is an important equity consideration.

How does Vibe Coding differ from traditional software engineering?

Vibe Coding is great for rapid prototyping and personal projects, which professional software engineers also do.

But I think a big difference concerns responsibility. When I think about my own preparation in computer science, there were reasons we needed to take courses on the social impact and ethics of computing. Vibe coders usually don’t need to concern themselves with the same kinds of questions as professional software engineering teams, such as reliability, safety, security, and maintainability.

Vibe is often optimized for coding How many wows can I get in the next hourRather for the quality of what is made or for those who can rely on it.

Vibe is often optimized for coding How many wows can I get in the next hourRather for the quality of what is made or for those who can rely on it.

As someone with a computer science background, how does your experience with Vibe Coding differ from someone who has never coded before?

Content knowledge has its advantages! You may have a different idea of ​​what is possible to create. You can more clearly describe what you want to create and more easily recognize when something has gone wrong and how to fix it. I think there are also habits of mind through discipline (ie, persistence and willingness to repeat) that feel especially helpful when coding vibes.

But more than the differences, what I noticed on the course was how much we had in common, regardless of one’s prior experience and skills. There is something deeply exhilarating about having an idea, bringing it into the world, and sharing it with others — and enjoying that wonder and joy together.

How do you see Vibe Coding moving forward?

I hope that more people have the opportunity to open up new opportunities to learn and express themselves creatively with Vibe Coding. I especially hope we see more of this in schools, although there will be challenges due to cost, resistance from lack of familiarity and understandable concerns about the potential impact on knowledge and critical thinking.

Vibe Coding doesn’t exist in a vacuum — its adoption in schools (and beyond) will be shaped by politics, policy, and people as much as the technology itself.

I think the central practices we develop while coding Vibe — thinking creatively about what we want to build, composing and iterating on prompts, critically evaluating what is built — are going to become central life practices. Maybe it’s less “vibe coding” and more “vibe everything”.

If the technologies around us are capable of doing an almost unlimited number of things for us, and we only need to know how to ask, then being able to imagine possibilities, articulate what we want to see in the world, review and iterate what we create will be incredibly helpful abilities for all of us to develop.

–

This is the story is reprinted with permission From the Harvard Gazette.

***