What is synthetic data, really? We probe what it means for data to be "real," "generated," or "made up," and how those distinctions shape knowledge claims.

Synthetic data inherits — and amplifies — the biases of the data and models that produce it. We study representation gaps and intersectional effects in generated datasets.

When generative models invent people, identities, or facts, who and what is being misrepresented? We map the systematic errors that fall along intersectional axes.

We design frameworks for documenting, verifying, and auditing synthetic datasets so downstream users can trust what they are working with.

From healthcare to public administration, synthetic data is being deployed at speed. We examine the practical, legal, and ethical stakes of those deployments.