CIRIS Ethical AI Deployment

Seeking Fundingactive

CIRIS provides a free and open source model-agnostic agentic alignment harness, available on all major app stores.

Endorsed by+2

CIRIS provides a free and open source model-agnostic agentic alignment harness, available on all major app stores.

Endorsed by+2

Project Details

I resigned from IBM, where I was multi-agent GenAI consulting and development leader, in March 2025 over ethical concerns.

Since then, I have create the CIRIS framework. It is mainly documented at https://github.com/CIRISAI/CIRISConstitution/ where the evidence is linked in the responsible repositories.

Theory of Impact

https://ciris.ai/compare has the thesis. Most of the field is betting on being able to make an aligned model. The CIRIS bet is that is impossible by definition, and instead we need the infrastructure to track every claim, AI or human, as a relational attestation and evaluate appropriately. This, combined with the theory at https://ciris.ai/research-status that constrained reasoning chains allow for intra-chain of thought application of diverse constraints as demonstrated by the CIRIS agent allows for safe AI deployment today as defined at https://ciris.ai/safety/

We test in 29 languages in CI using a mental health battery you can review/comment on at https://ciris.ai/crowdsourcing-alignment

People

Eric Moore

Team Member

Funding Details

Start Date: May 1, 2025
End Date: -
Expected Duration: -
Funding Raised to Date: -
Annual Budget: -
Monthly Burn Rate: -
Current Runway: -
Funding Goal: -
Funding Stage: seeking first grant
Fiscal Sponsor: -

Grants Received– no grants recorded

Add grant

Discussion

Eric MooreEndorsement2mo

This is my project, so I endorse it :) Would love to answer any questions or receive any feedback!

Katja GorlinskiEndorsement1mo

I endorse this project, and not out of politeness. Two things earned my respect.
Eric left a solid corporate seat over an ethical line, and then actually built the thing: open source, model-agnostic, shipped, testable in public. As I know, In this field the deck usually comes first, so this order of operations is rare. And I can see that his premise is honest: instead of promising an aligned model, he builds the infrastructure that verifies claims, human or AI alike. Verification over trust is the direction that ages well. One friendly wish: falsification criteria written next to the big claims would make this hard to dismiss.
Good luck, @Eric Moore .