Our work translates specific, well-defined problems into programs people can actually use. Each project is scoped to a concrete need, grounded in practice, and designed to deepen rather than broaden over time.
Flagship program
Flagship program. Active.
A workshop-based program helping young people use AI without outsourcing their thinking. Skeptia is built around real scenarios students already face and draws on intelligence-analysis frameworks — adapted for young audiences — to teach source literacy, verification habits, and independent judgment.
Learn more about Skeptia →Research track
Research track. Active.
As AI agents take on increasingly autonomous action, incidents involving misaligned behavior will become more common, and harder to investigate. Existing frameworks cover reporting obligations and causal-factor analysis reasonably well, but none address the investigative work that matters most once an incident is flagged.
Our research position is that intentional agentic misalignment is structurally analogous to intentional insider threat, a parallel the AI safety field has begun to recognize, and for which traditional investigations practice offers substantial, underused resources.
Read our research position →In development
Directions under consideration include educator training on AI-dependent learning patterns, responsible-adoption frameworks for small organizations, and resources addressing the use of AI as a substitute for human relationships — particularly among young people.
Get in touch →