Straight from the Source: How You’re Bridging the AI Effectiveness Gap

Artificial intelligence is no longer a future-facing experiment for the social impact sector. According to the Blackbaud Institute’s Bridging the AI Effectiveness Gap report, 85% of social impact professionals are already using AI at work. The challenge isn’t adoption. It’s turning that adoption into meaningful organizational impact. 

While AI use is widespread, a much smaller group of organizations have moved beyond individual productivity gains to create measurable organizational outcomes. These AI-adaptive organizations are distinguished by stronger governance, clearer policies, greater transparency, and a more intentional approach to data readiness. 

To better understand what that looks like in practice, we asked your peers from across the social impact sector to discuss the findings with each other. Their conversations revealed organizations wrestling with the same questions highlighted in the research: How do we adopt AI responsibly? How do we maintain trust? And how do we scale our impact without sacrificing the human relationships at the heart of our work?

Starting Small, Learning Fast

One of the clearest themes across sectors was that organizations aren’t looking for sweeping transformations overnight. They’re looking for manageable ways to experiment.

In higher education, one participant described their organization’s approach using a memorable analogy:

“Adopting AI in some way is like if you’re painting a room and you’re painting like a small little corner that nobody sees just so that you could see how it looks. And then if you like it, then you decide to paint the whole room.” 

That measured approach appeared repeatedly across sectors. Organizations are testing AI with content development, prospect research, donor segmentation, scheduling, and operational tasks before expanding into more constituent-facing applications.

The conversations highlighted a consistent pattern: leaders are looking for practical, low-risk opportunities to learn what works within their own organizations before scaling those efforts more broadly. 

Transparency Is No Longer Optional

One of the report’s most striking findings was the gap between donor expectations and organizational practices. While most donors want transparency about how organizations use AI, far fewer organizations currently provide it. 

That topic dominated discussions across nearly every breakout room.

Participants consistently returned to the question of trust. For many, transparency wasn’t simply about disclosing the use of AI. It was about helping donors, constituents, students, patients, and patrons understand how their information is being used and what safeguards exist to protect it.

As one higher education professional put it:

“I think it’s more probably like, if I give you information, how is AI going to use it?… What kind of guardrails [do] you have in place with the human in the loop? Those are, I think, the important things to document.” 

Others wrestled with a practical challenge many organizations now face: what exactly should be disclosed? If AI helps draft a fundraising appeal, assists with donor segmentation, or supports content creation, where is the line between useful assistance and disclosure-worthy use? 

While there was no consensus, there was broad agreement that organizations should proactively communicate how AI supports their work and where human oversight remains involved. Several participants discussed the need for AI policies, ethics guidelines, internal governance frameworks, and public-facing statements that help build trust before stakeholders start asking questions. 

Governance Is Emerging as a Competitive Advantage

The report highlights governance as a key differentiator between organizations that are merely using AI and those that are seeing measurable value from it. 

The discussions reinforced that finding.

Many organizations described being in the early stages of policy development. Others admitted they had no formal guidance in place yet.

One K-12 participant summarized the situation many institutions find themselves in today:

“Everybody’s doing their own thing.” 

Another emphasized that before organizations can create meaningful guardrails, they first need a shared understanding of the tools themselves:

“There just needs to be an understanding of what all of these actually do, how they collect and store data. And then that education needs to be shared with everybody.” 

Healthcare organizations provided examples of what more mature governance can look like. Participants described approved AI environments, internal AI councils, extensive security reviews, and policies designed to ensure compliance with privacy requirements. 

At the same time, practitioners acknowledged that creating policy can feel like trying to hit a moving target.

As one nonprofit leader observed:

“By the time [a policy is] published, some aspects of it are obsolete.” 

Even so, the consensus was clear. Organizations that are making progress are not waiting for perfect answers. They’re establishing principles, building governance structures, and creating room to learn as the technology evolves.

Data Readiness Still Matters

The report identifies data readiness as another factor separating AI-adaptive organizations from the rest. Organizations that have confidence in their data, understand where it resides, and have clear processes around its use are better positioned to translate AI adoption into meaningful outcomes. 

Practitioners repeatedly connected AI success to data quality, security, and stewardship.

Arts and culture leaders discussed conducting internal audits to better understand which AI tools are being used across their organizations and whether sensitive donor information could inadvertently be exposed through consumer-grade tools. 

Healthcare professionals emphasized enterprise-approved systems, oversight committees, and ensuring organizational data remains protected and separate from public model training.

K-12 and higher education participants spoke about balancing innovation with institutional responsibilities, privacy considerations, and the need for clear guidance around what information can and cannot be shared with AI systems. 

Together, these conversations reinforced a simple truth: getting value from AI starts with knowing and protecting your data.

The Human Element Remains Non-Negotiable

Perhaps the most compelling theme to emerge from the breakout sessions was that practitioners do not see AI and relationship-building as opposing forces. In fact, many participants believe AI’s greatest value lies in helping organizations spend more time on the work that only humans can do.

One arts and culture professional captured that sentiment perfectly:

“The whole point is actually so that you can continue to have that human connection with your donors and the people that you serve. You’re actually and can enhance that by taking things off your plate that have nothing to do with actually talking to donors, like digging through data.” 

That perspective surfaced repeatedly. Participants described using AI to draft communications, conduct prospect research, identify opportunities, and streamline administrative work while intentionally preserving donor-facing interactions as deeply human experiences. 

Some organizations have already established clear boundaries. One K-12 advancement professional explained that while their team uses AI to strengthen general communications and operational workflows, they continue to write highly personalized donor acknowledgements themselves because those messages are fundamentally relationship-driven. 

As organizations begin experimenting with more donor-facing applications of AI, transparency and human access remain key concerns. One nonprofit participant suggested a practical approach:

“If we were going to use AI to send out an email series or text or whatever it was going to be, we may want to just actually put it at the end of the message… and always provide them with a way to actually talk to a real person.” 

That recommendation encapsulates a broader theme running throughout both the research and the practitioner discussions. Stakeholders are increasingly comfortable with organizations using AI to improve efficiency, but they still expect authentic human connection when it matters most. 

The organizations bridging the AI effectiveness gap aren’t necessarily the ones using the most AI. They’re the ones using it with intention, supported by governance, guided by transparency, and focused on strengthening rather than replacing the relationships that define the social impact sector. The technology may be moving quickly, but the goal remains unchanged: helping organizations better serve their missions and the people who make that work possible. 

Start your journey towards effective, responsible AI with the free AI for Social Impact Certification Program.