AI EQUITY PROJECT 2024
The year we asked: what is at stake?
The first year of the AI Equity Project captured a sector full of cautious curiosity. AI was already showing up in nonprofit work—but readiness, governance, trust, and infrastructure had not caught up.
708
Nonprofits heard from
59%
Using AI
27%
Felt ready
6%
Had an AI policy
ORIGINS
Why the AI Equity Project began
By 2024, AI was already influencing how nonprofits wrote, communicated, analyzed information, made decisions, and imagined the future of their work. But the people closest to communities were rarely the ones shaping the rules, tools, or expectations around that adoption.
​
The AI Equity Project began as a research and listening initiative to understand that gap. We wanted to know not only whether nonprofits were using AI, but whether they felt prepared to use it responsibly, whether data equity was part of the conversation, where trust was breaking down, and what kinds of support organizations actually needed.
​
In the first year, 708 nonprofit organizations across the U.S. and Canada helped create a baseline for the sector. Their responses gave us a picture of AI at an early, messy stage: present, promising, unevenly understood, and largely unsupported by formal governance.
THE DATA
2024 at a glance
INDICATOR | 2024 |
|---|---|
Organizations participating | 708 |
Using AI | 59% |
Feeling ready for AI adoption | 27% |
Familiar with AI | 53% |
Familiar with AI bias | 44% |
Familiar with Data Equity | 50% |
Have Data Equity practices | 46% |
Have an AI policy | 6% |
Taken together, these numbers told us something important: AI adoption had already started, but organizational readiness and governance were still at the beginning.
FINDINGS
What did we learn?
01
AI was already here—but mostly informally
AI was not a future issue for nonprofits in 2024. Organizations were already experimenting, but much of that use was ad hoc and centered on relatively accessible tasks such as drafting content, brainstorming, and basic analysis. The technology was entering everyday work faster than shared organizational practices were being built around it.
That matters because informal experimentation can be useful for learning, but it can also leave decisions about privacy, data, bias, procurement, and community impact to individual staff members rather than the organization as a whole.
02
Adoption was moving faster than readiness
Nearly six in ten organizations reported using AI, while fewer than three in ten felt ready for adoption. Most respondents described their data and AI infrastructure as average to far below average, and uncertainty and mistrust remained widespread.
The gap was not simply about access to tools. Organizations were trying to learn while adopting—often without enough time, technical confidence, shared language, or examples of what responsible AI looked like in nonprofit settings.
03
Governance was barely beginning
Only 6% of organizations reported having an AI policy. That meant most experimentation was happening before clear expectations had been set for responsible use, human oversight, data protection, or accountability.
The 2024 findings made governance visible as more than a compliance exercise. For nonprofits, the absence of guardrails can affect how communities are represented, whose data are exposed, which biases are repeated, and who is responsible when an AI-assisted decision causes harm.
04
The sector needed people and capacity—not just technology
When nonprofits named what would help them move forward, leadership support, staff training, and technology tools rose to the top. The first two are telling: organizations were not simply asking for new software. They were asking for people inside the organization to understand the change, make decisions together, and build confidence around it.
The 2024 baseline pointed toward a central lesson that would grow stronger in later years: responsible AI adoption is as much an organizational and human-capacity challenge as it is a technology challenge.
EQUITY
The equity signal we could already see
In 2024, familiarity with AI was higher than familiarity with AI bias, while awareness of Data Equity was only beginning to translate into consistent practice. The project’s first AI Caution Ratio—measuring how many people familiar with AI were also familiar with its risks and biases—was 83%.
The question was not simply “Who gets access to AI?” It was also: Who gets enough information to question it? Who carries the consequences when it fails? And whose knowledge should shape what responsible adoption looks like?
LEGACY
Why 2024 still matters
2024 gave the AI Equity Project its baseline. It captured a moment when the sector was still asking foundational questions: What is AI? Where could it help? What could go wrong? Are we ready?
That baseline matters because later years show what changed—and what did not. Familiarity rose. Use expanded. Policies began to appear. But the original questions about power, trust, data, bias, readiness, and community voice did not disappear. They became more urgent as AI moved deeper into organizational practice.
CONTINUE THE STORY
From curiosity to accountability
The AI Equity Project is designed as a multi-year body of learning, not a one-time snapshot. Explore how the sector moved from cautious curiosity in 2024 to active experimentation in 2025—and toward questions of accountability, impact, and stewardship in 2026.