Best AI for AdvisorsWhat is data retention?
Short answer
How long a vendor or your firm keeps data (recordings, transcripts, prompts, outputs) and when it's deleted.
Definition
Data retention is the set of rules for how long information is kept and when it's destroyed. For AI tools, it covers recordings, transcripts, prompts, uploaded files, outputs and derived data like embeddings.
There are two sides: what the vendor keeps on its systems, and what your firm keeps to meet its own recordkeeping obligations.
Why it matters for advisors
The two sides can pull in different directions. Your firm may have to keep certain records for years under the books and records rules, while you'd prefer the vendor to keep client data as briefly as possible. The answer is often: the vendor deletes quickly, and your archive keeps what you're required to keep.
Amended Regulation S-P also expanded the disposal rule to cover customer information, so deletion at the end of a relationship matters.
Example
A firm sets its notetaker to delete audio after 30 days, exports final summaries to its archive, and confirms in the contract that all data is deleted within a set period after cancellation.
Watch out for
"Deleted" in the app may not mean deleted from backups. Ask vendors how long backups are kept.
Related terms
- Zero data retentionAn arrangement where an AI provider doesn't store your prompts or outputs after returning a response, beyond what it needs to process them.
- Books and records ruleRule 204-2 under the Advisers Act, which lists the records SEC-registered advisers must keep, including certain communications.
- Model training on customer dataWhen a vendor uses the data you put into its tool to improve or train its AI models, which may expose client information.
- EmbeddingA list of numbers that represents the meaning of text, letting software find related documents by meaning rather than exact words.
Learn more
Sources
Last updated .