Inside the Agentic Self: What ASU students build in will.i.am's class
The Agentic Self is a three-credit ASU course taught by will.i.am, founder and CEO of FYI.AI, that teaches students how to thoughtfully design and direct AI agents as part of creative, ethical and human-centered work.
In the course, you’ll build and refine a personal AI agent grounded in your voice, values, identity and purpose. Then, you’ll put that agent to work investigating a problem you choose and developing and testing a solution. You’ll prepare a three-minute pitch and submit it for the course Showcase. No coding background or technical major is required.
will.i.am has described the course’s larger purpose this way: “The Agentic Self represents a solution to AI replacing human jobs.”
The course meets live for three hours each Monday, with each session combining three distinct experiences: instruction led by will.i.am, conversations with invited guests from research, business and the arts who are putting agentic AI to work in their fields, and hands-on studio time with the teaching team to apply what you’re learning to your own agent and project.
Between live sessions, you’ll continue developing your agent and project, share work in progress, exchange ideas with classmates and receive feedback from your instructors. Your agent, your understanding of the problem and your solution will evolve together: what your agent helps you discover will shape what you build, while the needs of your project will shape how you configure, test and guide your agent.
Throughout the course, you’ll also consider privacy, authorship, bias, consent and the responsibilities that come with directing an AI agent.
The role an agent plays will look different for every student. For Dhruv Patel, a computer science student who took the course in spring 2026, that meant creating a partner for exploring and developing ideas: “I want my agent to be a riffing buddy or a friend I can bounce ideas off of,” Patel said. “I want it to think like I do, but with all the computational power and all the other intellectual capabilities that I don’t have behind it.”
How the course is structured: Four Pillars, Four Projects
The course is organized around four pillars that move from defining your agent to choosing a problem, developing a solution and making the case for it.
- Purpose & Agent (Weeks 0–1): What agent will I build, and what is it for?
Project 1: Agent Build - Problem (Weeks 2–3): What problem will I own, and why does it matter?
Project 2: Problem Brief - Solution (Weeks 4–5): What solution will I build, and how will it work?
Project 3: Solution Draft - Pitch (Weeks 6–8): Can I win support for it?
Project 4: Pitch Recording and Final Materials
The tool: EDU.FYI
At the center of the course is EDU.FYI — a new ASU and FYI.AI platform designed specifically for building and experimenting with agentic AI. It's where you'll do hands-on project work, in-class workshops and the development of your Agentic Self.
What students are building
You build your agent by putting it to work on a problem you care about. As you investigate the problem, test ideas and develop a solution, you’ll continually refine how your agent thinks, responds and works with you. And because the agent represents you, building it also means making deliberate choices about how it should operate. You’ll work through questions of authorship, consent, bias and privacy, documenting what your agent contributed, where it fell short, and the rules and safeguards you put in place to guide it.
That emphasis on agency and ownership is central to how students describe the experience.
“What I really like about the class is that he’s teaching us how to use AI to work on our behalf, because right now AI is owned by the big corporations that get to use your data and you don’t have any say,” said Jeremiah Holland, a graduate student in narrative and emerging media at ASU’s Los Angeles location. “With this class and with this AI system, the data is under our control and we get to decide how it’s used.”
Professor Peter Murrieta, one of the course’s co-instructors, describes that same idea through the lens of creative ownership.
“The idea of creating an agent for yourself where you own all the materials you’re putting into that agent makes it much easier to move forward in the world,” he said. “Anything I tell it, I own, and nobody’s scraping and learning off of it.”
That philosophy extends to how AI is used within the course itself. Each instructor is represented by a purpose-built AI agent that can interact with students and student-built agents throughout the semester, giving students another way to engage with course ideas and explore agent-to-agent collaboration. Human instructors remain fully responsible for instruction, assessment and academic oversight.
Meet the teaching team
will.i.am, faculty lead, is a globally recognized artist, producer, entrepreneur and technologist whose work spans music, media, education and artificial intelligence. As founder and CEO of FYI.AI, he leads the development of human-centered AI systems focused on creativity, learning and self-expression, and he brings that creator's perspective to the course's questions of identity, voice and responsible AI use.
Peter Murrieta co-instructs from Los Angeles as interim associate dean of the Herberger Institute for Design and the Arts and Deputy Director of the Sidney Poitier New American Film School. A multiple Emmy Award-winning writer and producer, his credits include "Mr. Iglesias," "Wizards of Waverly Place," and most recently, "Primo" on Amazon Prime.
Sean Hobson co-instructs across both Tempe and LA as chief design officer for ASU's EdPlus unit and an assistant clinical professor, where his teaching and research focus on agency, learning science and how people and institutions adapt during periods of rapid technological change. "Students aren't just studying the future, they're building it," he said of the course. "Learning by doing — that's the whole point."
Jorge Costa co-instructs from Tempe and online as an assistant professor of Studio Engineering and Recording at ASU. A recording engineer, producer, mixer and educator with more than 20 years of experience, his credits include work with artists such as Korn, Dr. Dre, Wiz Khalifa and Melissa Etheridge. A graduate of Berklee College of Music, he brings extensive experience in music production and engineering to the course.
Is this class right for you and how do you enroll?
The Agentic Self is designed for students across majors and levels of technical experience. No coding background is required. Instead, the project-based structure and collaborative learning community give you space to experiment, test ideas, revise your approach and build as you learn. If you’re curious about AI but don’t think of yourself as a “tech person,” this course was built with you in mind.
Because learning happens through collaboration and hands-on work, regular participation in the three-hour live session each Monday is an important part of the experience. The studio-based format brings together ASU students in person in Tempe, students joining in person from the Los Angeles area at the FYI Studio, and ASU Online students participating remotely — creating one learning community across locations and disciplines.
Enrollment is limited, so early registration matters. Current and prospective ASU students can register for the Spring 2027 session here.