If you have any questions or feedback, please email us at [email protected]!
- By default, Slipbox enforces Privacy Mode, which means zero data retention. None of your transcripts, summaries, notes and audio or derivatives of the above will ever be stored or used to train our AI models or any third party.
- If you choose to generate summaries or chat with Slipbox, your transcript is temporarily sent to our servers to generate the summary or chat response, and then deleted immediately after.
- Note that even without bringing your own AI keys, your data will not be stored or used to train our AI models or any third party. We are working on a feature for users to bring their own AI Keys.
- Slipbox collects critical telemetry and error data to improve and support the application.
Slipbox is one of the only AI note-taking tools that truly protects your privacy. We take a privacy first approach to protecting your data. We take a fundamentally different approach to protecting your privacy through our hybrid agent architecture. Instead of storing everything in the cloud like traditional meeting assistants, none of your sensitive data is stored in the cloud. Advanced features like Summaries, AI Cues, and Ask Slipbox prompts are sent to the cloud temporarily but its only used for a few seconds and then discarded.
When your information is sent to the cloud we use AES-256 encryption with TLS 1.3+ to ensure that your data is secure.
Frequently Asked Questions
How does Slipbox ensure my meeting data stays private?
Slipbox transcribes audio directly on your device. Unlike traditional cloud-based solutions, that sends your audio to a third party to be processed. Your audio data stays completely local, and our hybrid architecture only uses cloud processing when absolutely necessary - with immediate data disposal afterward. Think of it like having a personal AI assistant that lives on your Mac rather than in someone else's cloud.
Can I use Slipbox offline while maintaining privacy?
Yes! Most of Slipbox's core features work entirely offline thanks to our local-first approach. You can transcribe whatever and wherever you want, and your data stays local. Try this out by turning off your internet connection and transcribing a meeting.
How does Slipbox's approach compare to other AI note-taking tools?
Most AI note-taking tools follow a cloud-first approach, storing and processing everything on their servers. This creates privacy risks and often leads to compromises between functionality and security. Slipbox's hybrid architecture demonstrates that you can have both - powerful AI features and robust privacy protection. In a world where a couple seconds of your audio could be used to create a fake AI persona, we beleive that privacy should be a top priority.
What features needs my data to be processed in the cloud?
Currently, summaries, AI Cues, and Ask Slipbox prompts are processed in the cloud. The information is sent to a light agent in the cloud that coordinates requests to the model provider like OpenAI, Anthropic, or Google. None of the information is stored in the cloud or used to train AI models by Slipbox or any third party.
How It Works
Figure 8. High level overview of Slipbox's architecture
Local-First Processing
At the core of Slipbox's privacy approach is our local-first processing strategy. Your meeting transcripts and conversations never leave your device unless absolutely necessary. We use local agents powered by SLMs to decdie when data needs to be sent to the cloud. This approach ensures that your valuable meeting context stays under your control while still enabling powerful AI features.
Cloud Integration
When we do need cloud processing for more complex tasks, we take a unique approach. Unlike traditional cloud-based tools that store everything permanently, our servers act as lightweight coordinators. Any data that needs cloud processing is handled immediately and then discarded - think of it as a brief consultation rather than long-term storage. This means you get the benefits of powerful cloud AI when needed, without compromising your privacy.
Privacy Benefits
Our hybrid architecture delivers four key privacy advantages. First, you maintain complete control over your sensitive information since it primarily stays on your device. Second, by minimizing data transfer to the cloud, we significantly reduce potential privacy risks. Finally, our approach naturally protects team privacy by keeping sensitive discussions local, which is especially valuable for organizations handling confidential information.
Future Enhancements
We're actively working on expanding our privacy capabilities even further. We're also investing in more capable on-device processing and expanded local AI capabilities, always with the goal of keeping your sensitive information under your control while delivering powerful features. We are working on an end to end encrypted backup system to ensure your data is secure during storage and transfer.
As the models and capabilities improve, we will be able to offer more and more features fully offline on your device.
What does this mean for businesses?
As an IT admin, you may still want certain controls to ensure that your company data is secure. We offer admin console controls for Organization owners that can be applied to your organization. We also have options to backup data into your own cloud storage providers.
Please contact [email protected] to discuss your organization's needs. Some enterprise features include:
- Organization level data retention policies
- SAML/SSO authentication
- SCIM and Audit Logs
- Exporting Audit logs to your DLP or SIEM tool
- Automated backup to your own cloud storage providers