The missing layer

Most AI stacks combine three layers โ€” data platforms, application workflows, and model inference โ€” but lack a federation layer that coordinates model training, evaluation and analytics across organisational boundaries. MCP and Arazzo help with interoperability at inference and workflow time; federated learning frameworks solve interoperability at optimisation and governance time.

Why it matters for public services

Councils, NHS bodies and ALBs often need more than a tool-calling assistant: a controlled way for multiple data holders to contribute to shared intelligence without merging raw data โ€” implying client selection, aggregation, privacy controls, node identity, auditability and federation governance.

Relates to: Flower, Cross-device vs cross-silo, The missing collaborative-learning layer.

Terms

Glossary terms used here: Audit trail.