About

What Sathi is

Sathi is personal intelligence that grows around your life — your conversations, documents, calendar, family and devices, understood together, privately, on infrastructure you control. It began as one family's daily assistant and is built so that any person or family, anywhere, can eventually join or run one. Today it serves its first ring of real users in Nepali, English, Japanese, and Hindi.

Who builds it

Sathi is built and operated by its founder. The company behind it is Sathi Systems— entity detail, corporate documentation, and how the company earns live on the company's own site, sathisystems.com. Altruistic AI is not a company: it is an open architectural framework anyone may implement, and Sathi is its reference implementation.

Who owns what

The rule that everything else hangs off: you and your family own your data and the understanding derived from it; the operator owns the software that operates on it. Your ownership is mechanical, not rhetorical — total signed export, provable deletion, a front door that accepts your archive back, and a signed receipt when you leave. The framework Sathi implements is open; forks of the idea are expected and welcome.

Why is Sathi free for individuals?

Because individuals are not the product. There is no advertising, no engagement optimization, and personal data is never sold — that is architecture, not policy, and it does not change with a pricing page. Individuals and families use Sathi free today.

What is intended to sustain it is the company's business, stated in the same honest tenses on sathisystems.com: physical interfaces, sovereign infrastructure for organizations, and transparent referral partnerships — announced publicly by name, commissions disclosed, never advertising, user reviews published unedited, and a commission never reorders advice. None are live today. The free personal layer is the point of the system, not a loss leader waiting to be monetized.

Is Sathi open source?

The architecture is open: Altruistic AI is a published framework anyone may implement. Sathi's own source code is not published today. The openness that is guaranteed, now and permanently, is the one that protects you: your archive exports as signed, self-describing, readable files; paired devices are repointable; nothing about your exit depends on our permission. More may open over time — decided deliberately, not by drift.

If Sathi disappears

Continuity is a product requirement, not an afterthought: the system is built to survive its own company. Your export can rebuild a Sathi elsewhere; the framework is open; devices can be repointed. The full answer lives on /privacy.

What we publish, and what we don't

The rule: radical transparency about your rights and the system's behavior; deliberate reticence about competitive implementation.

• Published: the principles, the promises, the authority ladder, the data boundary and every provider that receives anything, the memory semantics, the capability inventory with honest statuses, the threat posture and its limits, and everything you can inspect, export, or delete.
• Not published: retrieval and ranking internals, prompts, optimization details, deployment automation, hardware designs, unreleased products, and exact roadmap sequencing. None of it affects your rights.