# AI *The «AI» section is intended for creating and editing artificial intelligence. This section displays all the AIs you have created, their types, and the tasks in which they are used.* ## General view of the section In the Selarti personal account go to the «[*AI*](https://selarti.com/panel?menu=ai)».
## Main elements The central part of the screen displays a list of all created AIs. * **Name** — the name of your AI; * **Type** — the selected type of communication («Incoming communication» or «Incoming and outgoing communication»); * **Used in** — lists the tasks in which the AI is used; * **Control** — buttons for copying and deleting the AI. ## Creating an AI Go to the «[*AI*](https://selarti.com/panel?menu=ai)» section, click «*Create AI*» in the top right.
A window for creating an AI will open, in which you need to fill in the fields: * **Name** — specify the AI name so you can identify it; * **Select GPT version** — from the dropdown list choose the version that fits your goals. We recommend GPT-4o. Read more about their differences in [the article](/manual-en/additional/ai-models.md); * **Work option** — choose one of two types of communication with the client: * «Incoming communication» for responding only to incoming messages; * «Incoming and outgoing communication» for sending outgoing messages and responding to incoming ones.
Additional settings for «Incoming and outgoing communication» When this item is selected an additional field will appear — «*Start message*», the initial phrase with which the AI will begin the dialogue with the client. For a productive dialogue we recommend starting it with a greeting and a brief explanation of why the AI agent is writing to the client. For VK, Telegram, WhatsApp channels we recommend creating variation in start messages so the account is not blocked for mass messaging. For this use [spintax](https://spintax.neytin.com/app#/) and the construction {Message 1|Message 2|Message 3}. This way the AI will randomly choose the message that will start the dialogue. **Example start phrases:** {Good afternoon{!|)|👋|😊} I found you in the Marketing for Business group. {Tell me, are you interested in discussing sales automation using AI|Tell me, is sales automation using AI of interest to you}?|Greetings{!|)|👋|😊} I found you in the Marketing for Business group.{Tell me, are you interested in discussing sales automation using AI|Tell me, is sales automation using AI of interest to you}?|Hello{!|)|👋|😊} I found you in the Marketing for Business group. {Tell me, are you interested in discussing sales automation using AI|Tell me, is sales automation using AI of interest to you}} For Iframe and Page Messenger channels, variation in dialogue start is not yet provided. However, you can make the dialogue start with three messages using the construction {Message 1&\&Message 2&\&Message 3}. These messages will be sent sequentially.
* **Response delay in seconds**— the time the AI will «think» after the client's message. If the client sends additional messages during this pause, the AI will process them in a single combined response within the same context, instead of replying to each message separately. This setting also allows AI agents to appear more human — it's hard to believe you're talking to a person if you receive a whole paragraph in a couple of seconds; * **Settings: creativity, diversity threshold, reasoning effort, and verbosity** — depending on the selected model, you may be able to configure one or more of these parameters (see the [article](/manual-en/main-sections-of-the-service/ai/settings-creativity-diversity-threshold-reasoning-effort-and-verbosity.md) for details). a) Creativity — adjusted with a slider, affects the AI’s ability to imagine and elaborate on responses. Excessively high creativity may cause AI «hallucinations», such as incoherent text, mixed languages, and other errors. We recommend a value of 0.5 if you are unsure which setting to choose. b) Diversity threshold — adjusted with a slider, affects the variety of AI responses. The higher the diversity threshold, the more unique the responses become, but excessively high values may lead to «hallucinations» If you are unsure which setting to choose, set it to 0.5. c) Reasoning effort — can take one of five values. The higher the reasoning effort, the more deeply the AI analyzes the request and the more accurate the response may be, but this can increase generation time. If you are unsure which setting to choose, use the medium value. d) Verbosity — can take one of three values. The higher the verbosity, the more detailed and longer the AI responses become, but excessively high values may make the text overloaded. If you are unsure which setting to choose, use the medium value. * **Instruction** — also called a prompt, this is the most important stage of creating an AI. Write the instruction as clearly and fully as possible in the language in which AI agents will communicate with clients. It is important to give the AI a clear technical task, define its role, task, experience, even its name. Before writing, read [the article](/manual-en/additional/how-to-create-instructions-for-ai.md) about writing instructions for AI. {% hint style="info" %} For some AI models the «Creativity» and «Diversity threshold» settings are not available; there are alternative additional model settings with similar functionality. More in the [article about models](/manual-en/additional/ai-models.md). {% endhint %} Using the «Click to upload a file» button make your model even smarter. You can upload up to 100 files with useful information that the AI can use in dialogues. For example you can upload: sales scripts, converted phone call transcripts, dialogues between managers and clients, presentations, knowledge bases. Supported formats: TXT, DOC, DOCX, PDF, HTML, MD. {% hint style="info" %} The minimum uploaded file size is 1 Kbit, the maximum is 50 MB. {% endhint %} After filling in the fields click «*Create AI*»*.*
### Tag settings Tags are needed so that the AI can automatically tag dialogues and you can easily find them in the personal account. Click on the created AI, and in the opened settings window on «*Professional settings*», to configure tags.
Select the required tag from the dropdown list, describe in which cases the AI will use it and click «*Add function*». More about setting up and using tags in [the article](/manual-en/main-sections-of-the-service/ai/ai-functions.md).
## **Testing the AI** Test the AI to ensure it is working correctly. To do this open the created AI, click «*Test AI*»*.* A dialogue with the AI will open on the right; in the top right corner of the dialogue there are three buttons — to share, clear and restart the dialogue and to close the testing mode. Write your question in the input field «*Write your message*» and click «*Send*»*.*
If you are not satisfied with the bot's answer, you can edit the instruction and click «*Save*», and then continue testing. ## Editing an AI To make changes to an existing AI click on it, make the necessary changes in the opened editing panel and click «*Save*».
--- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://manual.selarti.com/manual-en/main-sections-of-the-service/ai.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.