Project summary CIRIS is open-source infrastructure that makes autonomous AI accountable instead of asking you to trust it. Every decision is cryptographically signed and hash-chained, the architecture makes deception structurally expensive (the Coherence Ratchet), and a kill switch sits below cognition where a compromised agent cannot refuse it. It is AGPL-3.0 and legally mission-locked as an L3C so it cannot be captured or sold closed. It is already live on the App Store and Google Play in 29 languages. This grant funds the next milestone: taking the accountability layer from centralized to fully federated, so it runs peer-to-peer with no central point of control or capture.

What are this project's goals? How will you achieve them? The goal is accountability infrastructure that no single party, including me, can quietly switch off or close. Concretely, over the grant period: ship the CEWP/CEG wire format live across federated nodes so signed reasoning traces and trust state replicate peer-to-peer; bring the decentralized build (IPFS as bootstrap substrate) to the already-shipped app; and continue hardening the per-language mental-health safety batteries that currently gate every release. The architecture is specified and the centralized version is in production; this is execution from a working base, not research from zero.

How will this funding be used? Primarily my full-time engineering effort to ship the federation milestone, plus minimal infrastructure (federated node hosting, build and signing pipeline, device testing on low-end Android). I am the sole full-time developer; the dominant cost is focused builder time. At the $10,000 minimum I ship the core federation milestone: CEWP/CEG live across nodes with peer-to-peer replication of signed traces and trust state. At the $300,000 goal I sustain full-time development for a year, bring on a second engineer, fund an independent security audit of the cryptographic attestation and kill-switch path, and extend the decentralized build and 29-language safety hardening into the medical and GDPR verticals already in the codebase. Each increment between buys proportional runway and scope.

Who is on your team? What's your track record on similar projects? I am Eric Moore, founder of CIRIS. Before this I was an IBM Associate Partner, worked in AWS Professional Services, and maintained AG2. Over the past year I built CIRIS from concept to a deployed, multi-platform system in 29 languages, fully open across 30+ repositories. Esu (Ethiopia) field-tests the Amharic deployment on budget Android hardware; Haley leads outreach. As an external validity signal, an IETF working group recently and independently arrived at the same governance architecture I had been shipping, which I read as evidence the design is sound rather than idiosyncratic.

What are the most likely causes and outcomes if this project fails? The single largest risk is founder runway: I am solo and self-funded, and if I run out of room before the federation milestone ships, momentum stalls. Secondary risk is adoption, that the infrastructure is built but under-used. The mitigating structural fact: everything is AGPL and public, so even in the worst case the work is not lost. The specifications, the deployed code, and the formal verification persist for others to build on. This is failure that leaves a commons behind, not a deleted private effort.

How much money have you raised in the last 12 months, and from where? Effectively zero external funding. The work to date has been self-funded. No prior grants.