TURTL & AI

AI agents are only as good as the data they’re trained on. Hatch connects directly to your content engagement, pipeline and revenue data ingesting both structured and unstructured data from across multiple sources to contextualize it's responses. 

Over time, Hatch will continuously learn from your company’s live content, revenue intelligence, and GTM actions, automatically refining recommendations and actions.

The result is trustworthy, contextualized, high quality outputs that drive meaningful results from day one, and only improve over time.

No. Your content is processed securely and never used for training unless you explicitly opt in to train a bespoke image model.

All core AI services run on Turtl’s AWS infrastructure. This environment is already listed in our Terms of Service and security documents.

We use AWS for infrastructure, Replicate for optional media generation, and Anthropic Claude via API for translation. Replicate can be disabled if needed for compliance.

You do. All generated media outputs are yours for commercial use. The underlying AI models are licensed to Turtl and cannot be redistributed.

You upload approved images to your Turtl library. AI can match images to your content or generate new ones using your custom model.

Yes. You must own commercial rights to any images used to train a model.

Claude for translation, Flux via Replicate for images, and hosted LLMs for content manipulation. No models are trained on your data unless you opt in.