Is our data used to train the AI?
On business API tiers from the major providers, no — your data is not used to train their models by default, and enterprise agreements make that contractual rather than a checkbox.
That is different from consumer chat apps, where the terms are written for individuals rather than for you, and where an employee pasting a customer list has agreed to something on your behalf.
The distinction that matters
Someone asked me about a large bank using a "watered-down" version of a well-known assistant. What they were describing is an enterprise deployment with contractual data terms. It is not watered down. It is the same capability under an agreement written for businesses.
That option is available to you as well, and it is what I build on.
Where the real risk is
Not the provider. Your own team, using consumer tools on their own logins, because the official route is too slow or does not exist. Every business I have worked with has some version of this happening.
Banning it does not work. Providing a sanctioned alternative does.
What to actually do
- Use business or enterprise tiers, never consumer accounts, for anything touching customer data.
- Get the data processing agreement in place — with the provider and with me.
- Design so the model receives the minimum it needs. If it does not need names, do not send names.
- Give your team a sanctioned tool, or they will find their own.
Got a version of this problem? Twenty minutes minimum, seven questions, and a one-page Bottleneck Blueprint within 48 hours. If it's not a fit, we'll say so.
Book a call →