FAQ

ScamAI offers a real-time API that detects whether images are AI-generated or authentic. Our solution helps prevent scammers from using deepfake technology to deceive users. We use advanced machine learning models to analyze digital patterns and artifacts that are characteristic of AI-generated images.

With the rise of sophisticated AI image generation tools, it's becoming increasingly difficult to distinguish between real and fake images. This poses significant risks for online safety, identity verification, and content authenticity. Our detection service helps maintain trust in digital communications and prevents fraud.

Our API analyzes multiple layers of an image using advanced AI models trained on millions of examples. It examines digital signatures, pixel patterns, and other technical characteristics to identify AI-generated content. The analysis happens in real-time, providing quick and reliable results for high-volume applications.

Our API supports common image formats including JPEG, PNG Images should be at least 256x256 pixels for optimal results. While we can analyze most types of images, the service works best with photographs and realistic renderings. Maximum file size is 10MB per image.

Average response time is under 6s for standard images for higher resolution images it may take up to 15s. Response times may vary based on image size and complexity. Our enterprise plans include priority processing with even faster response times. We maintain a 99.9% uptime SLA for all paid plans.

Integration is straightforward with our RESTful API. Start by obtaining an API key from your dashboard, then use our SDK (available for Python, JavaScript, and Java) or make direct HTTP requests. We provide comprehensive documentation with code examples for common use cases and frameworks.

Yes, we provide official client libraries for major programming languages including JavaScript, Python, PHP, Ruby, and Java. These libraries handle authentication, request formatting, and error handling. We also maintain community-supported libraries for other languages.

Yes, our API is optimized for real-time use cases. You can integrate it into your upload flow to screen images before they're published, or analyze existing images in your content database. We provide webhooks for asynchronous processing of large batches.

Your API key should be treated as a sensitive credential. Store it securely in environment variables, never expose it in client-side code, and use restricted keys for different environments. You can generate and revoke keys from your dashboard at any time.

We prioritize data privacy and security. Images are processed in memory and are never permanently stored. All API traffic is encrypted using TLS 1.3, and we're compliant with GDPR and CCPA requirements. You can enable additional privacy features like automatic image redaction.

Yes, enterprise plans include additional security features such as IP whitelisting, custom data retention policies, audit logs, and dedicated infrastructure. We also support single sign-on (SSO) integration and can sign custom security agreements.

Yes, we offer a 14-day free trial with up to 1,000 API calls. No credit card is required to start. The trial includes access to all standard features, allowing you to fully evaluate the service before committing to a paid plan.

We offer flexible pricing tiers based on monthly API call volume: Starter (up to 10,000 calls/month), Professional (up to 100,000 calls/month), and Enterprise (custom volume). All paid plans include technical support, with additional features available at higher tiers.

If you reach your monthly limit, you can either upgrade to a higher tier or pay for additional calls at our overage rate. We'll notify you when you reach 80% of your limit. Enterprise plans include flexible scaling options to handle usage spikes.

Our current models achieve about 92% accuracy on known AI generators but we are working on improving this. We continuously update our models to detect new generation techniques and maintain accuracy. False positive rates are about 8% for standard use cases. Enterprise customers receive access to our highest accuracy models.

We release model updates bi-weekly to maintain high accuracy against new AI generation techniques. Critical updates for emerging threats are deployed faster. All updates are backward compatible and require no changes to your integration.

While no detection system is perfect, we employ multiple analysis techniques to maintain robust detection. Our models are regularly trained on attempted circumvention techniques. We offer bounty programs for researchers who identify ways to improve our detection capabilities.

All paid plans include email support with 24-hour response time. Professional plans add priority support and scheduled technical consultations. Enterprise plans include 24/7 support, dedicated account management, and custom SLAs. We also maintain extensive documentation and tutorials.

You can contact our support team through your dashboard, email support@scam.ai, or use our developer community forum. Professional and Enterprise customers can also schedule video calls with our technical team for complex integration questions.

Yes, our Professional and Enterprise plans include integration support. We can review your implementation, suggest optimizations, and help with best practices. Enterprise customers receive additional services like custom integration development and on-site training.