Bridging The Gap Between AI Interest And Readiness

Artificial intelligence (AI) is transforming the non-profit sector, moving from a phase of awareness-raising to a design phase where organizations are beginning to experiment with tools even before they feel fully ready. While AI offers potential benefits for workload reduction and communication, it also brings significant risks regarding data privacy, bias, and equity. Non-profits are currently navigating a tension between the pressure to adopt new technology and the need to do so ethically.

This report aims to understand the evolving trends, gaps, needs, and barriers related to AI adoption and equity in the U.S. and Canadian non-profit sectors.

Surveying The Landscape

Between May and July 2025, the authors collected data through a survey of 850 individuals working in or with non-profits across the two countries. This represents an increase in participation compared to the 2024 baseline, when 708 people responded. Around two-thirds of respondents were from the U.S., with the remaining one-third from Canada. The majority (55%) represented small organizations with fewer than 50 employees. Executive leadership roles were heavily represented, making up 43% of respondents.

The survey combined quantitative questions about usage and readiness with qualitative “Community Voices” to capture the nuance of personal experiences. Respondents were asked about their data equity practices, ethical concerns regarding AI behaviors, and funding challenges. Importantly, the study focused on AI equity, defined as the “ethical development, deployment, and use of AI systems that prioritize fairness, inclusivity, and justice, especially for historically marginalized or underserved communities.”

The Gap Between Usage And Readiness

The report reveals a sector stuck in “curiosity mode.” While 76% of non-profits reported using AI — a significant increase from 59% in 2024 — only 9% of respondents felt ready to adopt it responsibly. This disconnect highlights a critical risk: adoption is outpacing governance. Despite the surge in usage, only 7% of non-profits have an internal policy or guidance on AI use.

Several key themes emerged regarding this gap:

• Stalled readiness: Most organizations are at a beginner familiarity level with AI but recognize its strategic importance. Leadership fear or misunderstanding was frequently cited as a barrier, often correlating with low self-rated readiness.
• Infrastructure deficits: Many respondents, particularly from small organizations, lack basic data infrastructure. They often store sensitive data on individual computers without clear processes, raising significant security concerns.
• Dreaming with caution: When asked to “dream” about AI’s future, respondents overwhelmingly focused on fears regarding bias, misinformation, and data protection rather than aspirational scaling of their impact.

Furthermore, there’s a mismatch between what non-profits need and what they believe funders support. Technical assistance is the most desired resource, with 60% of organizations stating that they lack the in-house expertise to assess AI tools. However, many organizations struggle to pitch clear, actionable AI use cases to secure funding. In fact, 71% of respondents said they don’t have a clear use case to pitch to funders, and 57% believed that funders aren’t interested in AI. Consequently, only 4% of organizations have AI-specific training budgets.

The Equity Practice Drop

A critical finding is the disconnect between awareness and action regarding equity. The authors’ “AI Equity Awareness Index” shows that general knowledge about AI and bias has improved by roughly 38%. However, their “Equity Practice Ratio,” which measures whether organizations actually implement data equity practices, dropped from 92% in 2024 to 62% in 2025. This suggests that while non-profits are learning the language of equity, they lack the infrastructure to practice it.

Additionally, the report identified specific gaps based on demographics:

• Gender: Male respondents reported being “very familiar” with AI more often than female respondents (43% versus 34%), highlighting a confidence gap that could skew decision-making.
• Age: Leaders aged 45 and older reported much lower familiarity with data equity compared to younger peers aged 18 to 29.
• Organization size: Small organizations consistently cited difficulty measuring AI impact as a top challenge, pointing to a lack of evaluative capacity.
• Role: Self-employed and part-time workers often reported using AI without formal data infrastructure, increasing the risk of privacy leaks.

Limitations

The study acknowledges that the data isn’t bias-free and reflects the specific demographics of those who chose to participate. The heavy representation of small organizations and executive leadership may skew the perspective toward those with strategic oversight but limited operational resources. Furthermore, the decline in reported equity practices may reflect a more realistic and cautious self-assessment by organizations in 2025 compared to 2024, rather than an actual decrease in performance.

The Message For Advocates

The report moves beyond raising awareness to offering concrete action plans. For animal advocates in the non-profit space, the message is clear: readiness isn’t about racing ahead; it’s about building the “scaffolding” to use these tools safely.

There are a number of ways the non-profit sector can bridge the gap between AI interest and readiness:

• Invest in training: Prioritize staff training on AI and data equity basics. The findings show that this is one of the top support needs for adoption. Ensure this training reaches all levels, not just IT staff.
• Audit your data: Before adopting new AI tools, conduct a readiness audit. Ask how your team collects and stores data. If you’re storing sensitive information on personal drives, address that infrastructure gap before introducing AI.
• Seek collective support: The report proposes “Readiness Studios” — shared spaces for small non-profits to experiment with AI in low-risk environments. The authors also suggest developing an open-access resource library to democratize knowledge and reduce the confidence gap.
• Equity-first fellowships: Support leaders from underrepresented communities to become trusted AI guides.
• Ask the right questions: When evaluating a tool, don’t just ask what it can do. Ask: “What kind of system does this create?” and “Who will control the information that goes into it?“.

Overall, this report underscores that using AI tools isn’t enough: organizations must govern them. Rather than waiting for issues around bias to be fully solved, non-profits can embrace progress over perfection by engaging in responsible experimentation today.

This summary was drafted by a large language model (LLM) and closely edited by our Research Library Manager for clarity and accuracy. As per our AI policy, Faunalytics only uses LLMs to summarize very long reports (50+ pages) that are not appropriate to assign to volunteers, as well as studies that contain graphic descriptions of animal cruelty or animal industries. We remain committed to bringing you reliable data, which is why any AI-generated work will always be reviewed by a human.