Civic Chatbot Audit
Measures the accuracy of consumer AI chatbots on civic questions, grading every captured response against time-versioned ground truth.
Pipeline
Author
Define question templates, value sets, and targets — the axes of the audit matrix.
Ground truth
Import correct answers from official sources and confirm disagreements in the review queue.
Run
Compose a template into test cases, fan out a batch, and dispatch execution jobs to the runners.
Grade
The judge proposes a verdict against ground truth; a human confirms or overrides the final call.