As described earlier in relation to FIG. 1, different techniques may be used to generate a trust score. One example of generating a trust score includes applying a rule set to destination risk information, customer status information, and call history information. For example, a first rule may be used to evaluate destination risk information to determine a risk score that represents whether a destination endpoint is a high cost destination endpoint and/or whether the destination endpoint is associated with fraudulent activity. A second rule may be used to evaluate customer status information for a customer account to determine a monetary risk associated with allowing a connection request for the customer account to connect to a destination endpoint. A third rule may be used to evaluate call history information linked to the customer account to determine whether the customer account has a history of calling the destination endpoint, or similar destination endpoints. The trust score may be generated based on the results of the evaluations performed using the rule set. As a specific example, application of the rule set may determine that the destination endpoint is a high risk destination endpoint, the customer account is trusted to connect to high risk destination endpoints, and the customer has a history of connecting to high risk destination endpoints, and based on these results, a trust score may be generated to indicate that a connection request for the customer account may be allowed.