Teach AI what people are like, without holding a fact about anyone.
OMEM is memory for AI agents, but the reason it exists goes further. We want to give AI a real understanding of human nature and behaviour, learned as anonymous patterns that name no one. Those patterns live in your own intelligence bank, a file on your machine, and hold counts about behaviour rather than facts about people.
Learn people, without holding a person
Most of what AI knows about people it learned by reading them: their messages, their documents, the record of individual lives. OMEM learns the other way. From what it sees across many people it forms priors, regularities of the form "people who do X tend to do Y", and it stores them as counts, never as a fact about anyone.
The bank is yours, on your machine
Those priors collect in your intelligence bank, a file on your own computer. It holds counts over behaviour, not people. No name, no message, and no number about anyone can appear in it, because the door it enters refuses anything that could.
A corpus you can train on
The bank exports as a training set: one line per pattern, plain counts, under a permissive license. A model can learn the shape of how people behave from it and never hold a person. That is the objective: teach AI our nature from patterns people can stand behind, not from surveillance.
A prior yields to the person
A rate across a population is a place to start, not a verdict on the one in front of you. A prior fires only into a silence, and the instant that person's own evidence disagrees it steps aside. A general pattern never overrides an individual.
The bank grows as you use OMEM, and it stays on your own machine.
Run OMEM as your agent’s memory and it learns what people are like as it goes. Your data never leaves your computer unless you export it on purpose. Read the intelligence bank in the dashboard, and take the training set whenever you want it.