Getting started
How to use Storytell
Other Storytell products
About Storytell
How we manage Storytell user dataTroubleshooting
Our data-sanitization service -- Sparkle.ai
We've whipped up an open-source tool to ensure your data stays secure and protected. With data being passed to multiple AI systems like foundational large language models, speech-to-text and other data manipulation services, and vector databases, data privacy is especially critical. That's where Sparkle.ai steps in, offering a solution that allows you to use your private data without worrying about exposing sensitive information.
Sparkle.ai introduces a feature that sanitizes your data, replacing private and personal information with synthetic data. Whether it's your name, phone number, or email address, Sparkle AI ensures your sensitive details are automatically replaced, giving you peace of mind.
But that's not all! Sparkle.ai is an open source project and its code is available at https://github.com/ai-sparkle/sparkle. This means you can contribute, send PRs, and even peek at the code base. It's a community-driven initiative that values transparency and collaboration. Sparkle makes it accessible as a service.
The best part? Sparkle.ai is incredibly user-friendly. Simply upload your files, choose what you want to sanitize, and let Sparkle.ai do the rest. It's that easy! With Sparkle.ai, you can confidently use your data, knowing it's in safe hands.
Why would you want to use Sparkle.ai?
Why might you have a business need for Sparkle.ai? Before answering that question, let’s explore what might lead you down this path:
- Does your org face the challenge of information sprawl or data siloes?
- Do you lose institutional knowledge when folks off-board from your company?
- Do you have a resident expert everyone depends on?
- Are you trying to understand your customers better? A project you’re working on?
- Would your customers or employees benefit from an interactive FAQ? Process docs?
- Do you ever find yourself repeating yourself (i.e., potential users always ask these three questions)?
If you answered yes to any of these questions, creating a custom chatbot built on company data may be the solution you’re looking for. But here’s the catch: none of your sensitive data can leave the “building” — so how do you take advantage of the rise of conversational AI and respect your company policies?
We have an answer for you: Sparkle.ai. When you upload docs to Sparkle, it detects identifiable information and replaces it with synthetic data. When you download the sanitized version you’ll also receive a key showing what’s been replaced.
Once your data has gone through Sparkle.ai, it's safe for any AI, including Storytell. You can use Storytell to create custom chatbots or chat with any long, complex document via web or Chrome extension.
Why Storytell created Sparkle.ai as an OSS project to Sanitize Enterprise Data for AI use cases
What is Sparkle?
Sparkle is an open-source project that sanitizes sensitive data. You can learn how to use Sparkle here. Here's an example of what Sparkle does:
Why did Storytell create Sparkle?
Storytell.ai is a powerful AI system for bringing knowledge alive within an enterprise. We've heard from Fortune 500 users that in order for them to leverage the power of Storytell, they need to ensure their sensitive company data is not being sent to 3rd party systems like OpenAI (or other foundational LLM models), Zilliz (or other vector databases), Deepgram (or other processing services), etc.
How does Sparkle solve the "Enterprise AI data problem"?
We've built Sparkle to be a data sanitization pre-processing engine, and we've created it as a fully open source service. Open source was important to us so enterprises can fully inspect the code and even run it locally in their own environments. Sparkle is used to sanitize an enterprise's sensitive data before that data is sent to Storytell, meaning the actual sensitive underlying data stays safely within the enterprise and Storytell is only using swapped synthetic data in all it's services.
Here's how to use Sparkle.ai with Storytell.ai
- Our data-sanitization service -- Sparkle.ai
- Why would you want to use Sparkle.ai?
- Why Storytell created Sparkle.ai as an OSS project to Sanitize Enterprise Data for AI use cases
- Why did Storytell create Sparkle?
- How does Sparkle solve the "Enterprise AI data problem"?
- Here's how to use Sparkle.ai with Storytell.ai