The Ethicist Spotlight highlights people who are doing some of the most important work in our field. We'll get an insight into their career paths, learn what they are working on and understand their thoughts on the future of tech ethics.
I used to work on Watson doing the machine personality for an AI Tutor. That’s where initial questions came up for me in regards to transparency, explainability, and fairness towards the students that were interacting with it.
Who are you and what do you do?
My name is Milena Pribic and I work as an AI Designer at IBM, with a primary focus on best practices for ethical AI. I’m the global Design focal point for our AI Ethics Board so I contribute to scaling different AI Ethics projects and collaborations across the company and across different business units.
How did you first get involved in technology ethics?
I used to work on Watson doing the machine personality for an AI Tutor. That’s where initial questions came up for me in regards to transparency, explainability, and fairness towards the students that were interacting with it. There wasn’t guidelines specifically around ethics for design teams at IBM at the time so I ended up co-creating Everyday Ethics for AI. Since then, I’ve been expanding on the focus areas set out in that work.
I stress working directly with product teams and customers to make sure that our principles are actually making their way into what we create. It's crucial to measure success by how those efforts change not just the end product, but the process and behaviours of the AI team.
What are you working on at the moment?
I’m working on some cool stuff around explainability mechanisms for different personas as well as different exercises and workshops focused on positive behaviour change. I’m expanding some work on power dynamics in human/AI relationships as well as working with IEEE on establishing well-being metrics for AI.
_________________________________________________________________________
Want to keep up to date with the world of tech ethics?
_________________________________________________________________________
What do you see as the biggest challenge for technology ethics over the next 5 years?
Making sure that our AI systems are designed with the insights and participation of those who are most affected by them.
If you could go back in time, what work related advice would you give to yourself 5 years ago?
Give yourself enough space to work on all the “non-techy” things you love (like writing!) because they’re ultimately going to shape the way you approach communicating complex ideas.
How do you avoid ethics washing in your work?
I stress working directly with product teams and customers to make sure that our principles are actually making their way into what we create. It's crucial to measure success by how those efforts change not just the end product, but the process and behaviours of the AI team.
Whose work do you admire in this field?
In general, I really enjoy working directly with AI/HCI researchers to co-create design resources.
What ethics resources would you recommend to anyone in this field?
- IBM’s Team Essentials for AI course
- These ethics exercises over at designethically.com
- This interview on trusted AI ecosystems
- Subscribing to the Montreal AI Ethics Institute newsletter
- Artificial Unintelligence, Meredith Broussard
- Design Justice, Sasha Costanza-Chock
- Race after Technology, Ruha Benjamin
--
You can follow Milena Pribic on Twitter here.
Comentarios