If a customer is swearing would be good for an agent to have the choice to see the full message or not - with a AI summary of the message without the swearing.
ย
That way the agent can still warn the customer to not use the inappropriate language but is not having to see the language if they do not wish to.