Introducing 🌊LaVague, an open-source Large Action Model framework to automate automation.

Confidential AI Assistant - BlindChat

Leverage the power of conversational AI without compromising your data privacy. Our open-source, privacy-by-design approach ensures complete data confidentiality.

Secure Conversational Support

Being highly cost-effective for all businesses, LLM-powered AI Assistants provide personalized assistance and problem-solving capabilities without human intervention.
But concerns about privacy have caused some users to prefer traditional onsite support options over chatbots, hindering the full potential of LLMs in the industry.
Our solution ensures the confidentiality of personal information and transaction details even when using LLMs.

By unlocking the full potential of your AI assistant with BlindChat, your SaaS can deliver top-notch customer service, streamline operations, and increase user satisfaction.

Leveraging Llama 2 70B
model with best-in-class security standards

Loyalty Program Management

Root cause analysis of customer complaints

Legal search

Why BlindChat ?

BlindChat helps vendors guarantee they can be trusted with their users data, by offering a secure Confidential Computing solution that protects the privacy and security of SaaS offerings.

Lightweight Experience

Say goodbye to heavy bandwidth and computing constraints. Unlike the local version that downloads a 700MB model pull, this model is fully managed.

Optimal Performance

By serving state-of-the-art models, like Llama 2 70B, which traditionally wouldn't run on most devices, you can leverage GPT3.5-like models with privacy.

High Privacy

Thanks to the use of remote enclaves, your data remains end-to-end protected, not even our admins can see it.

How it works

Attested TLS Protection

BlindChat implements attested TLS to encrypt data during transmission, ensuring security and preventing tampering by confirming the server's identity.

Secure-Hardware-Based Technology

User data is decrypted and analyzed only in secure, isolated enclaves, safeguarding against unauthorized access, even by Mithril's staff, thus keeping sensitive information confidential.

Cryptographic Proof of Code Integrity

BlindChat offers cryptographic verification of its server code, enhancing transparency and trust in the code's security and authenticity.

Constant Innovation at BlindChat

BlindChat is continually evolving, introducing cutting-edge features and updates that enhance user experience and security.

Interactive Features with Local Documents

BlindChat's upcoming feature lets you securely search within your confidential documents, safeguarded by our specialized secure architecture.

Automated Hallucination Detection

Enhanced AI Reliability and Confidential Web Search. Focusing on improving AI response accuracy and offering a secure web search feature.

Join the community
Contribute to our project by opening issues and PRs.
Join the community, share your ideas, and talk with Mithril’s team.
Join the discussion
Contact us
We are happy to answer any questions you may have, and welcome suggestions.
Contact us