Data

What the model is trained and grounded on, and what that constrains. In scope: training-data supply and exhaustion, synthetic data, licensing and provenance, retrieval corpora, data quality effects on capability per token, contamination. Out: retrieval as a harness mechanism (harnesses), model architecture (models). NOTE: no live market call depends on this rung yet — it was added on Connor's direction, and gate 3 is currently unmet.