Łukasz Strumiński|Created: 22/06/2026|Updated: 22/06/2026Was this article helpful?
Introduction:
Provide exceptional customer service by supporting your agents with artificial intelligence. Experience the potential of Thulium AI assistant in customer interactions to achieve maximum productivity and high customer satisfaction.
Available AI assistant features:
Track unique types of interactions or events that are important to your organization using tag suggestions. Filter your analytics by tags and dive deeper into your company's results.
After clicking in the call details or using automatic call processing, we gain access to suggested tags related to a given call.
After using the Suggest tags option, a window will appear with up to five suggested tags:
By clicking the button , we confirm the selection of tags marked with a checkbox.
As part of the ticket process, select , then approve the suggested tags.
The process of adding tags suggested by the AI assistant should be saved in the ticket by clicking .
During a chat conversation, to receive tag suggestions from artificial intelligence, click .
AI uses the 50 most popular tags in a given system for suggestions.
Stop manually creating tags and focus on customer service.
Enable your agents to resolve customer issues faster by suggesting answers based on previously collected information. Use data from your knowledge base for this purpose and increase agent productivity with AI assistance.
The source of data for AI response suggestions is the knowledge base, specifically the articles contained therein. In order to make them available to the AI assistant, click on for a given article and select the appropriate option:
In the context of applying the articles themselves to artificial intelligence, it is recommended that:
They should concern a single issue.
The title of the article should indicate what it is about.
Articles should not be long, but as concise as possible.
Images in the article are not processed—only the text contained therein.
Links are also not processed.
Ticket
To initiate a response to a customer's question, click in the text editor. Then select:
The generated response can be prepared in the same language as the customer's message or in the system language set by the user.
The generated message content will appear in the dialog box:
In the dialog box, after clicking on the icon and selecting the Suggest a reply option, artificial intelligence will prepare a reply to the customer's message.
AI generates a response based on the customer's last message, but also takes into account the context of previous messages.
Artificial intelligence allows you to respond with a single click, ensuring that responses are timely and accurate.
Improve your messages by ensuring grammatical correctness and matching the style to your client. Tailor your communication to each client's situation.
After entering the message in the dialog box and clicking on in the text editor, a list of available options will appear, such as:
In order to make changes, e.g. stylistic changes, to the prepared message, click :
To improve the quality of an article in the knowledge base, click and select the text processing process you are interested in:
The AI assistant enables agents to provide truly high-quality responses in a fraction of the time.
Increase efficiency and eliminate repetitive processes with automatic post-call summaries. Let your agents focus on what they do best: serving customers.
After analyzing the conversation by AI Voice Analytics, an additional section will appear in the conversation details, i.e. Summary.
In the action menu for a given ticket, click to get a summary of it.
The generated summary can be copied to the clipboard by clicking .
A summary of the ongoing chat is available in the action menu after clicking .
Thanks to a concise explanation of the issue in the AI summary, agents can quickly familiarize themselves with the message in order to take effective action related to the issue.
Convert your knowledge base articles into realistic speech with a single click! Reach more listeners than ever before.
In the article, click and then select Generate article recording. After the text has been converted to speech, the Listen to article option will be available when you select
.
Save your agents' time by conveniently transferring knowledge aggregated in available articles via voice.