Enable Overage Billing for AI Tokens

Each month, your Oracle Analytics Cloud deployment includes an AI token allowance for using generative AI services. Users can exceed the monthly token allowance; however, additional token incurs charges. To allow token overage charges, you must configure an additional security policy on the compartment that hosts your Oracle Analytics Cloud instance. This policy allows Oracle Analytics Cloud users continue to use Generative AI after the monthly AI token allowance is exhausted. Oracle bills only for the tokens that exceed your monthly allowance.

After you authorize AI token overage, administrators can use the Generative AI page in Oracle Analytics Cloud to configure overage parameters, overage notifications, and restrict AI token overage usage. See Configure Token Terms and Billing Settings and Configure AI Token Usage Threshold Email Alerts.

For Oracle Cloud Infrastructure (OCI) Generative AI pricing information, see: OCI Price List

1. Verify that your tenancy is in an OCI Generative AI region.

   For a list of OCI regions that offer Generative AI services (specifically, xAI Grok models), see Generative AI Models by Region.

   If required, subscribe to a Generative AI region. See Subscribing to an Infrastructure Region.

2. Create a security policy on the compartment hosting Oracle Analytics Cloud that allows users to use Generative AI Chat and incur charges.
   1. In the OCI Console, select Identity & Security, then select Policies.
   1. Create a security policy similar to the following example:

      Allow any-user to use generative-ai-chat in compartment <compartment_name> where all {request.principal.id='<analytics_instance_ocid>'}

      Where <compartment_name> is the compartment that contains your Oracle Analytics Cloud instance.

3. Monitor usage and billing on the Cost Analysis page. See Analyze Costs for Oracle Analytics Cloud.