Zero data retention arrangements provide that submitted content is processed and discarded rather than stored. In AI contexts it typically means prompts and outputs are not retained after the request completes, and are not available for logging, review, or training.
Some vendors offer it as a configurable option and others only under negotiated enterprise terms.
Alternative Names:
Zero Retention, No Data Retention
Why it Matters?
Zero retention removes an entire category of risk: data that is not stored cannot be breached, subpoenaed from the vendor, or inadvertently used for model improvement. For legal work involving privileged material it is the strongest available protection, though it comes with tradeoffs, since retention supports debugging, abuse monitoring, and audit trails that a firm may separately need for supervision purposes.
Frequently Confused with
Related terms
Frequently asked questions
Is zero retention the same as no training?
What is the tradeoff?
