I am Asst. Professor of Cultural Data Analysis based in the Department of Media Studies and the Institute for Logic, Language, and Computation. I integrate computer vision, semantic technologies, and media and queer theory to both technically develop and critically examine AI systems used for multimodal cultural analytics.

My research explores how abstract social concepts—such as identity, toxicity, and vulnerability—are operationalized in, and shaped by, computational logic in the era of AI. I focus on human-centered and ethical AI, computational cultural analytics, cognitive-inspired visual sensemaking, emphasizing ethical and explainable AI methods for analyzing cultural (heritage) data.

Overall, I challenge and play with assumed binaries, like public/private, nature/culture, male/female, and white/colored. Overall, I am committed to producing knowledge that explores how human and non-human agents adapt within the dynamic interplay of technological advancement, societal values, and cultural diversity.  My overall aim is to seek new ways to think about power, representation, and technological change.

​Previously, I was a post-doctoral researcher at the Human-Centered Data-Analytics (HCDA) group at Centrum Wiskunde & Informatica (CWI), the national research institute for mathematics and computer science in the Netherlands. I have a Ph.D. in Computer Science and Engineering and a M.A. in Digital Humanities and Digital Knowledge from Università di Bologna (Italy). I have a Bachelor's degree in Human Evolutionary Biology, with a minor in Gender and Sexuality Studies, from Harvard University (U.S.A.).

Specifically, my current research directions include the following:

Identity Labels in Latent Spaces

• Investigating how Generative AI models learn and represent identity labels in their latent spaces, with a focus on detecting and mitigating biases and stereotypes.
• Analyzing the types of prompts, concepts, and labels people use to describe their identities and comparing them with labels assigned by AI.

Toxicity and Digital Literacy

• Developing AI models for detecting and analyzing toxicity in online platforms, with a focus on identifying patterns of toxic symbology in memes.
• Advocating for digital literacy skills, particularly in relation to online platforms and social media, to promote a more positive and respectful online culture.

AI and Vulnerable Data Subjects

• Using AI to analyze family vlog content, identifying patterns of potential exploitation to inform policymakers and the public about children as monetized data subjects.​
• Balancing public interest with individual rights, ensuring that the use of AI is ethically sound and respectful of vulnerable data subjects, such as children.