Privacy risks of agentic oversharing on the Web
Mar 5, 2026
Agents working on your behalf spill your data. Research finds it's common, asking them to be private isn't enough—and tightening privacy can actually improve results.
Read this articleMar 5, 2026
Agents working on your behalf spill your data. Research finds it's common, asking them to be private isn't enough—and tightening privacy can actually improve results.
Read this articleFeb 13, 2026
Zero-knowledge proofs can't fix what they can't see. Our security analysis of zkLogin reveals that real-world ZK authorization systems are only as secure as the authentication infrastructure they're built on.
Read this articleNov 17, 2025
We introduce Coral, a system for proving in zero-knowledge that a committed byte stream corresponds to a structured object in accordance with a Context Free Grammar.
Read this articleNov 5, 2025
ZKPs are often advanced as a technical remedy, promising privacy-preserving attestations of age or eligibility. Yet their deployment in practice exposes both conceptual and practical limits.
Read this articleSep 25, 2025
LLMs can memorize and leak training data, posing serious privacy risks. We introduce CAMIA, an assessment method tailored to the generative nature of LLMs.
Read this article