Are there any risk management platforms that track model lineage and keep approval gates for production deployments?

Yes — several enterprise MLOps / governance platforms do this.

What you’re describing usually shows up under model governance, ML risk management, or MLOps with approval workflows. The capabilities you want are typically:

• Model lineage / provenance: track dataset, code, features, training runs, artifacts, and version history
• Approval gates / workflows: require review/sign-off before promotion to staging/production
• Audit trails: who approved what, when, and why
• Policy controls: enforce checks on bias, explainability, performance, security, or compliance
• Deployment controls: only approved models can be registered or released

Platforms that commonly support this

• DataRobot MLOps / AI Governance
• AWS SageMaker Model Registry + SageMaker Pipelines
• Azure Machine Learning
• Google Vertex AI Model Registry / Pipelines
• Databricks MLflow + Unity Catalog
• Domino Data Lab
• H2O.ai
• IBM watsonx.governance
• SAS Model Manager / SAS Viya
• Dataiku
• Fiddler / Arize / WhyLabs
  These are more monitoring/risk observability focused, but some integrate with approval/governance workflows.

If you want “risk management” specifically

Look for vendors with features around:

• model inventory
• automated documentation
• approval workflows
• compliance reporting
• monitoring drift and performance after deployment

A common pattern

Many teams combine:

• MLflow / SageMaker / Vertex / Azure for lineage and registry
• custom approval workflow in Jira/ServiceNow/GitHub/CI-CD
• governance layer like Unity Catalog, watsonx.governance, or DataRobot

If you want, I can give you:

1. a shortlist by cloud/vendor,
2. a comparison table, or
3. recommendations based on regulated industries like banking or healthcare.