A series of student essays from the Berkman Klein Center at Harvard University argues that foundation models transform personal data into inferential capabilities, enabling AI systems to generate sensitive inferences about individuals from information never explicitly disclosed. For example, a model can aggregate purchasing behavior, social media activity, and conversational patterns to make reliable predictions about an individual’s mental health status, political affiliation, or income level. This challenges conventional privacy frameworks built around discrete, identifiable records, as the relevant concern becomes a model's ability to generate sensitive inferences that users cannot reasonably foresee or control. The article notes that while recent federal data privacy proposals have expanded notions of covered data to include inferred data, they continue to conceptualize privacy harms as arising from identifiable pieces of information. To protect individual autonomy and limit informational power asymmetries, the authors argue privacy regulation must evolve beyond governing data alone to govern inferential capabilities through expanded definitions of covered data, capability-rooted governance, and enforceable impact assessments. The article also references the Federal Trade Commission (FTC), which noted in 2021 that 'neutral' AI systems can produce discriminatory outcomes along lines of race and other protected classes, such as healthcare algorithms that worsen disparities for people of color.