AI memory
An AI assistant with memory you can see and control.
Sevoria uses selective memory for durable preferences and recurring context. Saved memories are visible in Settings rather than hidden behind a personalised answer.
What memory is for
Memory helps Sevoria retain useful facts such as your preferred answer length, the name of an ongoing project, or a recurring writing style. That means fewer repeated explanations and more consistent help across conversations.
It is intentionally different from chat history. History preserves conversations for you to revisit; selective memory stores a smaller set of facts that may help future replies.
User controls
- See saved memories in Settings.
- Delete individual memories.
- Clear all memories.
- Turn memory off completely.
Examples of useful AI memories
- “I prefer concise answers with the conclusion first.”
- “My current project is a portfolio website for architects.”
- “Use British English when editing my work.”
- “I am learning Python and understand basic functions.”
What should never become an AI memory
Do not save passwords, card numbers, private keys, medical records, confidential workplace information, or another person's private data. Memory is a convenience feature, not secure secret storage.
Why controllable memory matters
Personalisation is only useful when it stays accurate. A visible memory list lets you correct old assumptions and remove context that no longer applies. You can also turn memory off when you want every conversation to begin without saved preferences.
AI memory FAQ
Does Sevoria remember every message?
No. Selective memory is designed for a smaller set of durable facts and preferences; it is not the same as storing every transcript as memory.
Can I use Sevoria without memory?
Yes. Memory can be disabled, and guest mode is available when you do not need saved personalisation.
Where can I inspect saved memories?
Signed-in users can review and manage them in Settings. See the Trust Center and Privacy Policy for related data controls.