How mission-driven
organizations work
with artificial intelligence.
I study how foundations, nonprofits, and city governments adopt artificial intelligence. Different sectors, the same four questions: what the strategy is, who governs it, whether staff can judge what it produces, and whether the result still serves the mission.
I came to these questions through fifteen years studying racialized and stigmatized communities, and Muslim-led nonprofits in particular. That work kept returning the same finding: institutions misread the people they serve because the data and categories they rely on were never built to see them. AI does not fix that problem. It automates it, and at speed.
I am an Assistant Professor of Public Administration at Portland State University's Mark O. Hatfield School of Government, a Fulbright Scholar, and a Fellow at the Indiana University Lilly Family School of Philanthropy, where I study the use of AI in philanthropic advising. I also work with city and state agencies, foundations, and nonprofits to see how these ideas hold up in practice.
My work
The same questions, across sectors
A foundation deciding how to screen grant applications, a nonprofit drafting client communications, and a city answering resident questions after hours are not facing different problems. They are facing the same one, under different names and different accountability structures.
AI strategy
How should an organization decide where AI belongs? Strategy means thinking past individual tools to how AI fits the mission, the people, and the work — which problems are appropriate for AI at all, where it creates real value, where to be cautious, and how to tell whether it is improving anything. A city weighing a resident-facing chatbot and a foundation weighing an application screener are running the same calculation, with different people bearing the cost of getting it wrong.
The objective is not more AI adoption. It is better alignment between AI, organizational purpose, and the people an organization serves.
AI governance
As AI becomes part of everyday work, questions of responsibility, oversight, and risk follow. I study how organizations build structures that let them benefit from AI while keeping accountability and human oversight intact — including the parts public agencies cannot avoid, like procurement, records, and answering for a decision in public.
Where does AI belong, how should it be used, and who remains responsible for the outcomes?
AI literacy in organizations
AI literacy is an organizational capability, not an individual technical skill. It goes well past writing a good prompt: understanding how these systems work and where they fail, evaluating what they produce, and knowing when they should not be used at all.
Different roles need different literacy. A caseworker using AI daily, a manager reviewing AI-supported work, a department head setting policy, and a council or board providing oversight do not need the same competencies — which is why I study how to assess literacy across roles rather than training everyone the same way.
Where the three questions meet
Strategy, governance, and literacy are not separate problems. They converge on one question: when AI enters a decision, do the results still serve the mission and the people the organization exists to serve? A city answers that to its residents, a foundation to its grantees, a nonprofit to the community it was built for — different accountability, identical question.
I use the idea of cointelligence to study how decisions are actually produced — through interaction between human reasoning and machine output, rather than by one replacing the other. A public values framework then asks whether those decisions hold up against fairness, accountability, transparency, and mission fidelity. Cointelligence explains how a decision was made; the public values framework asks whether it was worth making that way.
This comes directly out of my earlier research. Studying Muslim-led and racialized nonprofits meant studying what happens when measurement instruments were built around a different population — giving that goes uncounted, organizations read as suspicious rather than ordinary, communities whose civic life does not register in the categories available. Those were equity failures produced by data, long before AI entered the picture.
AI inherits all of it. Systems optimize for what is measurable, not what a mission requires; reliance on AI erodes an organization's capacity to reason independently; and policy choices get embedded in systems that merely look technical. Each failure lands hardest on the people with least room to absorb it.
Four public values
- Fairness
- Accountability
- Transparency
- Mission fidelity
Together these frameworks support work that is honest, grounded, and accessible — for organizations, for leaders, and for the public they answer to.
The book
How Mission-Driven Organizations Work with AI: A Cointelligence Framework
A structured approach for organizations trying to figure out where AI can genuinely help, where it introduces risk, and how to build governance that accounts for both.
The framework gives leaders a way to evaluate AI tools against their own values and build the capacity to govern AI adoption over time — not just react to it. It treats the strategy, governance, and literacy questions as one connected problem rather than three separate initiatives.
What it covers
- Where AI genuinely helps
- Where it introduces risk
- Dividing work between people and AI
- Evaluating tools against organizational values
- Building governance capacity over time
- Who remains responsible for outcomes
A periodic briefing
What developments in AI actually mean for public, nonprofit, and philanthropic organizations — written for people doing the work, not for the technology press.
Research, practical tools, and cases on AI strategy, governance, literacy, and equity. No product announcements.
Free · Periodic · Your address is never shared
Topics
- AI strategy
- AI governance
- AI literacy
- Public values and equity
Get in touch