Soda Contract Copilot

Learn how Soda Contract Copilot works and what privacy & security measures Soda takes with AI language models.

Soda Contract Copilot is an AI-assisted feature that helps teams create, understand, and maintain Soda data contracts more efficiently. It acts as an intelligent guide throughout the contract lifecycle, making it easier to define expectations, detect gaps, and keep contracts aligned with real-world data behavior, without ever accessing or processing your raw data.

Contract Copilot is designed to support both:

  • Business and analytics users, who need assistance drafting data contracts and establishing data expectations, and

  • Data engineers and platform teams, who need precise, maintainable, and scalable contracts.


Enable/Disable Contract Copilot

Soda Contract Copilot can be configured from the Organization Settings.

  1. Click on the profile icon on the top right corner → Navigate to Organization Settings.

  2. Under the Organization tab > Soda AI: Check/Uncheck the "Enable Contract Copilot powered by OpenAI" checkbox to enable/disable the feature.

  3. Click Save on the top right corner.


Soda's AI security & privacy measures

Soda Contract Copilot is built with privacy and security as first-class design principles. The following safeguards apply to all AI/ML-powered features in Soda, including Contract Copilot.

Metadata-Only Principle

Soda’s AI/ML features exclusively process and analyze metadata and statistical summaries of your data; never actual raw data or record-level values.

Privacy by Design

Source data, such as row-level data and/or PII, is never sent to Soda’s AI/ML engine or to any external model provider by default.

No Customer Data Used to Train External Models

Soda does not use customer data, metadata, or results from your environment to train or fine-tune external AI, LLM, or ML models.

Diagnostics Data Remains Local

If record-level diagnostics (such as Failed Rows) are enabled:

  • They are stored only in your own data warehouse(s) via Soda’s Diagnostics Warehouse

  • They are not stored in Soda Cloud

  • They are not accessible to Soda’s AI/ML processing

Secure Encryption and Compliance

All AI/ML processing is performed securely and in alignment with Soda’s SOC 2 Type 2 compliance:

  • Data in transit is encrypted using industry standards (TLS)

  • Data at rest is encrypted using industry standards (AES-256)


Contract Copilot uses AI/ML techniques to help you write data contracts, analyzing:

  • Dataset schemas

  • Non-identifiable column-level metadata

  • Statistical summaries

  • Existing contract definitions and checks

Contract Copilot does not analyze or inspect:

  • Raw table data

  • Individual records

  • Row-level values

  • PII or sensitive fields

Soda Contract Copilot focuses on assisting decision-making, not replacing human ownership of contracts. The output is a set of suggestions, explanations, or guidance that you can review and apply.


You are not logged in to Soda and are viewing the default public documentation. Learn more about Documentation access & licensing.

Last updated

Was this helpful?