How does Lumo maintain the context of the full conversation?

Messages are processed by the LLM to generate responses, but they aren't stored indefinitely in plaintext. When you continue a conversation, the full conversation history is sent to the server (encrypted in transit) so the LLM has context. After processing, the conversation is stored with zero-access encryption at rest. The server doesn't retain plaintext copies beyond what's necessary to generate the immediate response.

AI that keeps your conversations private

Learn how Lumo's privacy and security architecture lets you research, query, and explore without putting your personal information at risk.

Chat with Lumo

What are the risks of mainstream AI chatbots?

Many popular AI services turn your conversations into data that can be sold, shared, or used to train their models. What feels like a private chat today may resurface in future AI responses, be fed to advertisers or governments to target you, or be exposed in a future data breach.

Lumo is an alternative AI assistant that helps you get more done while keeping your data protected.

How Lumo keeps AI chats private and secure

Encryption in transit and at rest

Lumo uses two layers of encryption to protect your messages both during and after your conversations, making it far more secure than other popular AI assistants. With Lumo, only you can decrypt your chats — not even Proton can read them.

Protecting your messages as they’re sent

When you send a message to Lumo, it's encrypted in transit while traveling between your device and Proton's servers. The server decrypts the message temporarily to process it through the language model, generates a response, encrypts it again, and sends it back. This protects your data from interception over the network.

No logs policy

Lumo doesn’t keep records of when you chatted, where you were, or what you discussed during your conversations. While most AI services collect and retain this metadata, Lumo doesn’t store it at any point.

No data training

Lumo never uses your conversations to train or improve its models. Unlike other AI assistants, your chats are not analyzed to build knowledge and are never at risk of reappearing as part of Lumo's response to another user.

Ghost mode

Switching on "Ghost Mode" ensures your current chat disappears completely as soon as you close it. Once the session ends, nothing is saved, synced, or stored anywhere, so you can chat with total privacy and confidence.

Based in Europe

Lumo runs on servers controlled entirely by Proton in Europe, meaning your data is protected by strong European privacy laws — not subject to US surveillance requests or laws like the CLOUD Act.

Chat with Lumo Get Lumo for Business

Why use a privacy-first AI assistant?

Protect propriety information

Brainstorm freely with Lumo on product development, competitive analysis, or financial planning without worrying about ideas leaking to competitors or ending up in someone else's Lumo conversation.

Secure sensitive details

Discuss life plans or confidential client work with confidence that your conversations are never logged, so your personal and professional vulnerabilities stay private — not stored, analyzed, or monetized.

Explore without exploitation

Research anything: from controversial topics to gift ideas without building a profile that follows you around the web. No targeted ads, no data selling, no algorithmic manipulation. Just genuine exploration.

Lumo's privacy and security model

The architecture behind Lumo’s privacy-first design and the features that make it a trustworthy choice for private, encrypted AI conversations.

What are the AI privacy concerns?

Here’s what’s at stake when AI has access to your personal data and what you can do to protect your privacy online.

We need private AI to prevent surveillance capitalism

If we want an AI that is not shaped by Big Tech or authoritarian politics, we need something different.

AI chats only you can read

Draft emails, solve technical problems, and find the answers you need knowing your conversations stay private.

Chat with Lumo

Frequently asked questions

How does Lumo maintain the context of the full conversation?: Messages are processed by the LLM to generate responses, but they aren't stored indefinitely in plaintext. When you continue a conversation, the full conversation history is sent to the server (encrypted in transit) so the LLM has context. After processing, the conversation is stored with zero-access encryption at rest. The server doesn't retain plaintext copies beyond what's necessary to generate the immediate response.