Faculty Spotlight In Conversation
Mihaela Vorvoreanu's work sits at the forefront of responsible AI, where she studies how these systems are designed, how they affect people, and how organizations can build them more thoughtfully. Currently engaged as a principal applied scientist at Microsoft, Vorvoreanu also teaches in the University of Washington’s interdisciplinary Master of Science in Technology Innovation (MSTI), the flagship degree at UW’s Global Innovation Exchange. Before Microsoft, Vorvoreanu had spent more than fifteen years in academia, developing and delivering cutting-edge user experience-focused programs at the intersection of communication, design, and technology. Read on for an edited Q&A with a faculty leader at UW’s Global Innovation Exchange:
Tell us about your work at Microsoft.
Fundamentally, I’m a responsible AI researcher, and I do research projects that are targeted toward supporting and advancing the practice of responsible AI.
This can mean different things. Early on, I worked on guidelines for human-AI interaction, which became what we call the HAX toolkit—the Human AI Experiences toolkit. It’s basically a set of tools that help people think through how AI features should behave when they interact with people.
A lot of my work is about understanding the people who build AI—how they work, what their constraints are—so that whatever guidance or tooling we give them fits into their workflows and their mindset.
What does your background allow you to see that others might miss?
It’s very easy to get excited by what technology can do and the fact that you can do something and you can build something without necessarily asking the questions: why? Who needs this? How is it going to affect people? Should we even build this?
I think people are much more interesting and more complex than computers. Computers are able to do a lot of things. But the critical thinking and the kinds of questions that you learn to ask when you have a liberal arts education—I think those are indispensable.
In liberal arts, you’re taught to question a lot. And I think that’s something that’s really, really important and often missing.
How has your experience at Microsoft informed your perspective on how things work?
In academia, we like parsimony. We have these theories and frameworks and processes that are very, very elegant. They can be represented neatly in a diagram on paper.
And I think what you see in industry is that real life and real decision making is awfully, awfully complicated and messy. And the diagrams and the processes seem naive and disconnected.
So, coming back to teaching, I think that’s perhaps the most important thing that I can bring. Not really diagrams and processes and theories, but more values that can help anchor a point of view and inform decision making.
Because at some point in your career, you quit working alone at your desk, or even just building things, and you start making decisions. And that decision making is talking to people. It’s messy. It has all sorts of points of view informed by all sorts of backgrounds. And to participate in decision-making you need to be able to know where you stand and what your values are.
What are students typically missing when they enter the tech sector?
If you’re not able to hold your own in a conversation, a meeting, or a debate, or if you fail to persuade others and explain your point of view, the rest of what you know is not necessarily going to help you advance.
At some point, [your work] becomes about participating in decision making. And that means being able to articulate your stance and influence others.
We’ve seen forever that students don’t have these—people call them soft skills, I hate that word—but the people skills. And I think even in how I used to teach before, I underestimated the importance of being able to influence others in conversations where there are multiple perspectives in the room.
How will your teaching change because of that?
I plan to do a lot more discussion and debate. Because at the end of the day, you have to be able to use your knowledge in conversation with others.
You’re very rarely going to just write a beautiful paper and be done. Yes, presentations matter, but in the workplace it’s about being clear in conversations where people might disagree.
So, I think adjusting the communication style to what I actually see happening in the workplace and thinking a lot about the importance of influencing other people—that’s a big shift for me.
What do you want students to take away from your classes?
I think the best thing that I can do is instill a set of values and principles that can guide decision making in the future.
I was going to start with fairness. That’s how I drafted my syllabus. And then last week I realized I’m going to start with human autonomy and agency, because I think in this day and age that might be the most important thing.
There are times when we happily give away our autonomy—to the dishwasher, for example. But once you start delegating your critical thinking or your ethical decision making, maybe you’re going to end up being less human than you are right now.
What mindset should students bring into this field?
I think a helpful mindset is that of a lifelong learner. Somebody who’s curious, who’s not afraid to ask difficult questions.
The most important skill that you get out of a graduate program is not really the specific skills—it’s how to learn. That’s the skill that carries you throughout a long career.
The world changes very, very fast. Technology changes very, very fast. I don’t really care to teach technical skills, because those get outdated in two weeks. But learning how to learn and being grounded in values: that’s what lasts.
How do you make the case for responsible AI in practice?
At the end of the day, if you’re harming too many of your customers, there’s not going to be anybody left to buy your product. Responsible AI is good AI.
I sometimes explain it like this: say you want to build a car, and you want to build it fast, but you’re in such a hurry that you don’t have time to put in a steering wheel. That would be completely unacceptable in the automotive industry.
And yet with AI, it’s somehow fine to ship something that doesn’t yet work properly, because it’s important to get it out there fast. That’s the tension.
How can people who don’t see themselves reflected in traditional tech culture still influence the direction of the field?
I think there’s power in numbers, in critical mass. There’s power in finding community of like-minded people.
It’s really, really hard to do this alone. I think if you’re alone, your voice can get lost. But once there is a critical mass and some form of organizing, of collective resistance, that’s when things start to shift. Also, there’s power in policy – we are beginning to see AI regulation emerge, and that can be influenced through voting and public participation.
Some days I might feel that this fight for responsible AI is not a fight we can win. But I am convinced that it’s a fight worth fighting.
Vorvoreanu’s combination of academic rigor, deep industry experience, and an interdisciplinary lens is central to the MSTI experience at the University of Washington’s Global Innovation Exchange. Her perspective reflects the kind of learning the program is built around: not just technical skill-building, but the ability to navigate complex, messy problems that don’t fit neatly into a single domain. Students here learn from faculty who are actively shaping the field and who bring both theory and practice into the same conversation—preparing them to move beyond where they are today and into more influential technology roles in the future.
Back to News
