> ## Documentation Index
> Fetch the complete documentation index at: https://docs.omnifact.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Interacting with AI

> Master the art of AI communication with practical examples and interactive exercises.

export const Glossary = ({children, term}) => {
  const glossaryDefinitions = {
    'ai': 'Artificial Intelligence - computer systems that can perform tasks typically requiring human intelligence',
    'artificial-intelligence-ai': 'Computer systems that can perform tasks typically requiring human intelligence',
    'ai-assistant': 'An AI system designed to help users with tasks through conversation',
    'space': 'A dedicated environment with a specialized AI assistant and optional Uploaded Files',
    'spaces': 'Dedicated environments with specialized AI assistants and optional Uploaded Files',
    'knowledge': 'The sub-menu in Space settings where you manage Uploaded Files and Connected Sources',
    'uploaded-files': 'Documents manually uploaded to a Space under the Knowledge tab',
    'prompt': 'Instructions or questions you give to an AI assistant',
    'response': 'The AI assistant\'s answer to your prompt or question',
    'query': 'A question or request for information',
    'context': 'Background information that helps the AI understand your request',
    'llm': 'Large Language Model - an AI system trained on vast amounts of text data to understand and generate human language',
    'large-language-model-llm': 'An AI system trained on vast amounts of text data to understand and generate human language',
    'privacy-filter': 'A system that automatically detects and masks sensitive information before it reaches AI models, and restricts images from being sent to non-EU hosted models for compliance',
    'chat-instructions': 'Custom guidelines that define how an AI assistant behaves and responds in a Space',
    'pattern-matching': 'The process by which AI recognizes and uses patterns in data to provide responses',
    'patterns': 'Recognizable structures or trends in data that AI systems can identify and use',
    'training-data': 'The text and information used to teach an AI system how to understand and respond to requests',
    'natural-language': 'Everyday human language that people use to communicate, as opposed to computer code or formal syntax',
    'hallucination': 'When an AI generates information that sounds plausible but is actually incorrect or fabricated',
    'filtering': 'The process of automatically detecting and temporarily replacing sensitive information to protect it during AI processing',
    'masking': 'Temporarily replacing sensitive information with generic labels to protect it from being seen by external AI models',
    're-insertion': 'The automatic process of putting original sensitive information back into AI responses after they\'ve been generated with filtered data',
    'retrieval': 'The process of finding and accessing relevant documents from Uploaded Files to answer questions',
    'hosting': 'Where an AI model is deployed and made available for use - either by third-party providers, on your organization\'s own servers, or in private cloud environments',
    'cite': 'To reference or mention the source of information, helping you know where an AI\'s answer came from',
    'sidebar': 'A panel that appears on the side of the screen showing additional information and options',
    'toggle': 'A button or control that switches something on or off, like opening and closing a sidebar',
    'navigation': 'The system of menus and buttons that help you move between different parts of Omnifact',
    'favoriting': 'Marking a Space as a favorite to pin it to the main navigation for quick access',
    'pinning': 'Attaching a Space to the main navigation bar so it\'s always visible and easily accessible'
  };
  const definition = glossaryDefinitions[term];
  return <Tooltip tip={definition}>{children}</Tooltip>;
};

## Master AI Communication

Talking to AI is a skill—one that pays off in better, faster results. This guide moves beyond basic prompting to help you understand the nuance of effective AI collaboration.

<Card title="The Goal" icon="bullseye" horizontal>
  Learn to craft requests that get you exactly what you need, the first time.
</Card>

## The Power of Context

<CardGroup cols={2}>
  <Card title="❌ Weak Prompt">
    **Example:** *"Write some documentation."*

    **Why it's weak:** The <Glossary term="prompt">prompt</Glossary> lacks essential details that the <Glossary term="ai">AI</Glossary> needs to help effectively. Without context about the type of documentation, intended audience, purpose, or requirements, the AI can only provide generic, unhelpful responses. It's like asking someone to "write something" without telling them what it's for or who will read it.
  </Card>

  <Card title="✅ Strong Prompt">
    **Example:** *"Write documentation for our new team onboarding process. The audience is new employees joining our marketing team. Include: daily responsibilities, key tools we use, and communication guidelines. Use friendly, clear language and provide examples where helpful."*

    **Why it's strong:** Clear audience, purpose, and requirements specified.
  </Card>
</CardGroup>

<Info>
  **Privacy is Built-In:** As you provide context, remember that Omnifact's **Privacy Filter** operates in the background. It automatically detects and masks sensitive information (like names, emails, and financial data) in your prompts and uploaded files before they reach the AI model, then intelligently re-inserts the original data into the response. You can write detailed prompts without worrying about data leakage.
</Info>

## Choosing How the AI Responds

Omnifact allows you to control how the AI processes your request through **Response Modes**. You can select the best mode for your current task using the selector below the chat input:

* **Auto (Default):** The AI automatically decides whether to use tools like web search, image generation, or document retrieval based on your prompt. It's the most capable mode for general use.
* **Quick Response:** The AI answers directly without using any external tools, providing the fastest possible response. Perfect for simple questions or brainstorming.
* **Simple Lookup:** (Available in Spaces with Uploaded Files) The AI strictly searches your uploaded documents and answers based *only* on that content, without relying on general knowledge.

You can also switch the response mode on any AI message and **regenerate** the answer to compare how different modes handle the same question. Learn more in [Response Modes](/en/platform/core-features/chats/response-modes).

## Sharing Files and Images

You are not limited to text. Click the **Plus** icon to the left of the chat input to attach documents, images, or audio files directly to your message. The AI can analyze, summarize, and answer questions about what you share.

* **In any chat:** Attach up to 5 files per message to get insights from reports, contracts, charts, or screenshots on the spot.
* **In Spaces:** Combine attachments with your Space's **Uploaded Files** for deeper, organization-specific context.

<Tip>
  Be specific about what you want from a file. Instead of "look at this," try "Summarize the key risks in section 3 of this contract."
</Tip>

## Regenerating Responses

When a response misses the mark, you do not need to retype your question. Use Omnifact's built-in controls to get a fresh answer:

* **Regenerate Response:** On the last AI message, click **Regenerate** to request a completely new answer to the same question.
* **Stop generating:** While the AI is still writing, click **Stop** to halt a response that is heading in the wrong direction, then regenerate or send a follow-up.
* **Switch response mode:** Change the mode on a previous message and regenerate to see how **Auto**, **Quick Response**, or **Simple Lookup** would answer differently.

Each regeneration produces an independent response, so results may vary. Treat the output as a draft and refine from there.

## Managing Long Conversations

Every AI model has a maximum memory limit called its **context window**. This is how much of the current conversation the model can keep in mind at once.

As a chat grows very long, Omnifact may show a warning that you are approaching the limit. When that happens:

* The AI may start to "forget" the earliest messages in the chat.
* Response quality, speed, or context awareness can drop.
* For best results, [start a new chat](/en/platform/core-features/chats/starting-new-chat) and carry over any essential context in your first message.

<Tip>
  Keep one topic per chat. When you switch subjects, start fresh rather than stretching a single conversation across unrelated tasks.
</Tip>

## Iterative Refinement (The "Drafting" Process)

**Breaking the "Google Search" Habit:** New users often treat AI like a search engine: they ask one question, get a mediocre answer, and assume the AI is "dumb" or "broken."

**The Truth:** It is normal for the first answer to be imperfect. Think of the AI as a **junior employee or a drafter**. You rarely accept the first draft of a report from a junior employee without comments; treat the AI the same way.

Instead of giving up, tell it what to fix. Think of AI interaction as a conversation, not a one-off query. Here's a real example:

<Steps>
  <Step title="Initial Request">
    "Write a blog post about productivity tips for remote work."

    **AI Response:** Provides a basic outline covering common topics like dedicated workspace, time management, and communication tools.

    **Why Refine:** The initial content is useful but too general and lacks specific, actionable advice.
  </Step>

  <Step title="First Refinement">
    "That's a good start. Can you focus specifically on maintaining work-life balance and include examples from different industries?"

    **AI Response:** Adds targeted examples like:

    * Tech companies using flexible hours
    * Healthcare professionals doing remote consultations
    * Sales teams managing virtual client relationships

    **Why Refine:** The content is now more specific but needs a more engaging presentation style.
  </Step>

  <Step title="Final Polish">
    "Great content! Now please adjust the tone to be more conversational and add some practical daily routines people can follow."

    **AI Response:** Transforms the content with:

    * Friendly, relatable language
    * Step-by-step morning and evening routines
    * Real-world examples from remote workers
    * Actionable tips readers can implement immediately

    **Result:** A polished blog post that's both informative and engaging.
  </Step>
</Steps>

## Troubleshooting Guide

When things go wrong, try these fixes:

<AccordionGroup>
  <Accordion title="Problem: Responses are too generic" icon="filter">
    **Solution:** Add constraints and specifics

    ```
    Please make this more specific by:
    1. Including industry-specific examples
    2. Adding concrete numbers/metrics
    3. Referencing real use cases
    ```
  </Accordion>

  <Accordion title="Problem: Missing key information" icon="circle-info">
    **Solution:** Use the gap-filling technique

    ```
    The response is missing information about:
    1. [specific gap]
    2. [specific gap]
    Please fill these gaps while maintaining the existing context.
    ```
  </Accordion>

  <Accordion title="Problem: Wrong tone or style" icon="pen">
    **Solution:** Use tone matching

    ```
    Please rewrite this in a [formal/casual/technical] tone.
    Reference tone: [provide example of desired tone]
    Key elements to maintain: [list important points]
    ```
  </Accordion>
</AccordionGroup>

<Note>
  Remember: Great AI communication is a skill that improves with practice. Start with these beginner practices and adapt them to your unique needs!
</Note>

## Next Steps

Now that you understand how to interact effectively with AI, explore these topics to learn more:

* Review [Understanding AI Assistants](/en/guides/ai-fundamentals/understanding-assistants) to better understand how AI helps your work
* Learn about [AI Capabilities and Limitations](/en/guides/ai-fundamentals/capabilities-and-limitations) to set realistic expectations
* Explore [How AI Works](/en/guides/ai-fundamentals/how-ai-works) to understand the technology behind the interactions
* Learn [Prompt Engineering](/en/guides/prompt-engineering/introduction) best practices to learn how to write more effective instructions for AI
* Get started with [Omnifact Quickstart](/en/platform/overview/quickstart) to put these skills into practice