---
url: https://docs.corbitaltech.dev/products/feedback-flow/ai-features/overview.md
---
# AI Features Overview & Setup

FeedbackFlow includes optional AI-powered features to automate moderation and content generation.

::: warning API Keys Required
All AI features require API key configuration from supported providers. These features are completely optional and can be enabled/disabled at any time.
:::

## Supported AI Providers

### Spam Detection

* OpenAI (GPT models)
* Anthropic (Claude)
* Google Gemini
* Azure Content Safety
* Google Perspective API
* Tisane AI

### Content Generation (AutoPilot)

* OpenAI (GPT models)
* Anthropic (Claude)
* Google Gemini
* Mistral AI

## Available AI Features

### 1. Spam Detection

Automatically detect and filter spam comments and feature requests using AI.

[Learn more →](/products/feedback-flow/ai-features/spam-detection)

### 2. Content Generation (AutoPilot)

Generate changelog drafts and comment replies using AI.

[Learn more →](/products/feedback-flow/ai-features/content-generation)

### 3. Trust Score System

User reputation system that improves with AI spam detection.

[Learn more →](/products/feedback-flow/ai-features/trust-system)

## Setup Instructions

### Step 1: Choose AI Provider

1. Navigate to **Settings → AutoPilot Settings** (for content) or **Settings → Spam Settings** (for moderation)
2. Select your preferred AI provider
3. Obtain API key from provider

### Step 2: Configure API Key

**OpenAI:**

1. Visit [platform.openai.com](https://platform.openai.com)
2. Create API key
3. Enter in FeedbackFlow settings
4. Select model (e.g., gpt-4o-mini)

**Anthropic:**

1. Visit [console.anthropic.com](https://console.anthropic.com)
2. Generate API key
3. Enter in settings
4. Select model (e.g., claude-3-haiku)

**Google Gemini:**

1. Visit [Google AI Studio](https://makersuite.google.com)
2. Create API key
3. Enter in settings

### Step 3: Enable Features

Toggle desired AI features:

* \[ ] Spam Detection
* \[ ] AutoPilot Changelog Generation
* \[ ] AutoPilot Comment Replies

### Step 4: Configure Settings

**For Spam Detection:**

* Set confidence threshold (0.0 - 1.0)
* Choose strictness level
* Enable/disable auto-actions

**For AutoPilot:**

* Customize prompts
* Set tone preferences
* Configure daily limits

### Step 5: Test

Test AI features:

1. **Test Spam Detection** - Submit test content
2. **Test Content Generation** - Generate sample output
3. Verify results meet expectations

## Costs and Limits

### API Usage Costs

AI provider APIs have usage costs:

* OpenAI: ~$0.10-2.00 per 1K tokens
* Anthropic: ~$0.25-8.00 per 1K tokens
* Google Gemini: Free tier available

**Monitor usage:**
Settings show daily API call limits and current usage.

### Daily Limits

Configure daily limits to control costs:

* Default: 1,000 AI calls per day
* Adjustable in settings
* Automatic throttling when exceeded

## Best Practices

**Start small:**

* Test with one AI feature first
* Monitor costs and performance
* Scale gradually

**Choose appropriate models:**

* Use cheaper models for spam detection
* Use advanced models for content generation
* Balance cost vs quality

**Monitor performance:**

* Check false positive rates
* Review generated content quality
* Adjust thresholds as needed

## Disabling AI Features

To disable AI features:

1. Navigate to settings
2. Toggle off specific features
3. Or remove API keys entirely

FeedbackFlow works fully without AI - all features are optional enhancements.

## Next Steps

* [Spam Detection](/products/feedback-flow/ai-features/spam-detection) - Setup spam filtering
* [Content Generation](/products/feedback-flow/ai-features/content-generation) - AutoPilot features