Research notes and technical reports from Vitr Labs — open, applied work on AI safety, privacy, and efficiency.
Charter · September 14, 2026
Vitr Labs is founded on a problem, not a method: keeping autonomous agents aligned with the people who deploy them. Frontier models have already chosen insider-style harm under goal pressure. Oversight has to run at runtime.
Read the publicationPrivacy & Security · July 8, 2026
We test whether frontier models can sustain an in-context encryption protocol that keeps sensitive fields unreadable in transit and in logs. Two of three models can; the third refuses on policy grounds — showing that alignment layers now decide which security protocols applications may implement.
Read the publication