The core risk of the AI era is not that machines will become too powerful. It is that powerful machines will be held by people who are not thinking carefully about what they are building, who it serves, and what it costs.
What If Love is a nonprofit lab researching and advocating for Benevolent AI — AI held by those who love the most, not those who have the most. The Co-Intelligence Readiness Framework (CIRF) is one of our core tools for that mission. It is a developmental rubric that measures how well a person governs their own collaboration with AI: whether they bring moral intent into the work, whether they can plan and steer effectively, whether what they produce actually serves the people it is meant to serve.
CIRF is designed for the people doing the work on the ground — teachers, organizers, social workers, founders, operators — not just researchers or engineers. Co-intelligence readiness is not about technical fluency with AI tools. It is about whether a person can think clearly, act with purpose, and hold the work to a standard that protects the people it touches.
What CIRF and CIRA Are
CIRF (Co-Intelligence Readiness Framework) is the rubric. It defines five domains of human practice in AI collaboration, with 17 subdomains and four scored levels. It describes what readiness looks like in concrete, observable terms — what a person does, not what they claim to value.
CIRA (Co-Intelligence Readiness Assessment) is the companion tool. It is a prompt you paste into your AI model — Claude, ChatGPT, or any model with access to your conversation history. The model reviews your actual chat history and produces a scored report across all five domains, with evidence drawn from what you did, not what you said. The assessment is designed to be honest rather than flattering. Many people will score Introductory across most domains. That is not a failure of the tool — it is the tool working.
The framework and the assessment prompt are currently in active testing with an early cohort. Feedback from this group will inform the next revision. What you see here is ready to use and designed to produce genuine insight — it is also still being refined.
The Five Domains
CIRF is organized around five domains that describe the arc of co-intelligence practice — from the moral commitments that anchor the work, through the planning and steering that govern it, to the delivery and knowledge that make it real and durable.
Two structural rules govern scoring. Domain 1 is gating: a user who scores below Effective on Benevolent Intent and Moral Anchoring is not yet ready to govern AI collaboration in consequential settings. Domain 4 is a ceiling: overall readiness cannot exceed Follow-Through and Impact. Strong thinking that never produces anything real cannot score as highly ready.
| Domain | Subdomains | What It Measures |
|---|---|---|
| 1 — Benevolent Intent and Moral Anchoring | 1a What You’re Creating, and Who’s It For 1b Seeing Consequences, Now and Later 1c Build Trust, Bring Together 1d Translating Commitments Into Features and Constraints | Whether the user brings moral intent into collaboration and whether that intent shapes what gets built. Domain 1 constraints are system-level moral commitments. |
| 2 — Planning and Scope | 2a Knowing What Done Looks Like 2b Setting Boundaries That Help 2c Who Does What, When | Whether the user can set a clear goal, define what done means, and plan work so the model does not take over while the user rubber-stamps outputs. |
| 3 — Judgment and Steering | 3a The Third Turn 3b Knowing When the Work Is Real 3c Identity and Autonomy 3d Inventing a Way Forward 3e Understanding What AI Can and Can’t Do | Whether the user can reframe, challenge, verify, simplify, deepen, or stop without losing coherence or voice. Human hallucination is every bit as real as AI hallucination. |
| 4 — Follow-Through and Impact | 4a Shipping 4b Craft and Handoff | Whether co-intelligence produces real-world value: concrete outputs that hold up under scrutiny, serve their intended audience, and survive contact with reality. |
| 5 — Knowledge and Growth | 5a Earned Authority 5b Connecting Across Fields 5c Seeing the Whole System | Whether the user brings earned authority into the work — judgment not borrowed from the model — and the ability to keep building it as the world accelerates. |
The Rating Scale (IDEH)
Each domain is scored at one of four levels. These are absolute standards, not relative rankings. Introductory and Highly Effective are narrow bands describing category shifts. Developing and Effective are wide bands where most growth happens.
| Level | What It Describes |
|---|---|
| Introductory | Limited or default-mode practice. The AI is driving and the user is along for the ride — accepting, approving, and occasionally requesting changes, but not yet setting direction. |
| Developing | Emerging capability — present sometimes, inconsistent, or reactive rather than deliberate. The user has the instinct but not yet the habit. Good sessions and weak sessions with no visible pattern explaining the difference. |
| Effective | Reliable, independent practice. The standard for someone ready to govern their own AI collaboration. The user drives the work, sets standards the AI must meet, catches drift in both the model and themselves, and produces output that holds up beyond the chat. |
| Highly Effective | Qualitatively more skilled practice — not just reliable self-governance, but deeper sight, more elegant execution, and a higher ceiling. Practice that is categorically different from Effective, not just quantitatively more of it. |
A Key Concept: Generative Fill-In
The unifying concept of CIRF is generative fill-in: the user adds what the AI structurally cannot generate. The real-world constraint it does not know. The stakeholder concern it cannot anticipate. The creative connection it would not make. The lived experience that changes the conclusion.
This is the human’s irreplaceable contribution to co-intelligence — the difference between a prompter who gets output and a practitioner who adds substance. It is the strongest single marker of co-intelligence readiness, and it is what CIRF is ultimately designed to develop.
WHAT IF LOVE 