Why AI Success in Higher Education Starts With Data, Not Models

Artificial intelligence is quickly becoming part of daily operations across higher education. Yet many university foundations and advancement organizations start by asking which AI tool to buy.

But a better question is whether their data is ready.

Foundation finance teams need to move faster, provide accurate reporting, maintain compliance with donor restrictions and support stewardship efforts with limited resources. Yet critical information often lives in PDFs, scanned gift agreements, emails, spreadsheets and systems that weren’t designed to work together.

The challenge turning disconnected information into reliable insight. And that reality is shaping which AI projects succeed and which stall before they deliver meaningful value.

The Data Readiness Challenge Is Larger Than Most Institutions Realize

The higher education community continues to move AI higher on its strategic agenda. According to the 2025 EDUCAUSE AI Landscape Study, 57% of respondents said AI is now considered a strategic priority at their institution, up from 49% the previous year. The study also found growing attention to AI governance, policies, and workforce readiness.

Those numbers show clearly that AI is no longer viewed as experimental technology. It’s becoming an operational consideration.

Yet many foundation and advancement offices still deal with fragmented information:

• Gift agreements may exist as scanned documents.
• Fund restrictions might be stored in PDFs.
• Financial information resides across foundation systems, university ERP platforms, procurement tools and donor databases.

As a result, staff members often spend significant time locating information before they can even begin analyzing it.

Generative AI performs best when it can access accurate, organized and trusted information. If source data is incomplete, inconsistent or inaccessible, even sophisticated tools will struggle to produce dependable results.

Organizations that achieve early success with AI often start by addressing foundational data challenges first. They improve document accessibility, organize historical records, establish data ownership and identify trusted sources of information. Those efforts create an environment in which AI can produce meaningful business outcomes.

Practical AI Use Cases Are Emerging Across Foundation Finance

Despite the challenges, many institutions are already identifying valuable applications for generative AI. And the most successful examples tend to focus on repeatable processes that consume significant staff time.

Gift restriction analysis is one example. AI tools can help review historical gift agreements, summarize donor intent, identify potentially restrictive language and support compliance reviews. Staff members still make final decisions, but the initial analysis can happen much faster.

Fund matching is another promising area. Internal stakeholders often struggle to identify which funds can support a specific initiative. AI can review fund documentation and provide summaries that help connect institutional needs with appropriate funding sources.

Reconciliations and exception management also offer opportunities. Finance teams regularly review large volumes of transactions and supporting documentation. AI-assisted analysis can help identify inconsistencies, reveal exceptions and organize information for review.

Many institutions are also using AI to draft first-pass reporting narratives, stewardship communications and internal summaries. While human review is still essential, generating an initial draft can reduce administrative burden and free staff to focus on higher-value work.

The common thread across these use cases is straightforward: AI works best when it supports people rather than replaces them.

Governance Determines Whether AI Creates Confidence or Concern

As institutions move beyond experimentation, governance becomes increasingly important. That’s why many AI discussions begin with trust. For example, can the technology be trusted? Will the output be accurate? And could it introduce risk?

Those are reasonable questions. But every AI use case carries a different level of risk. A draft meeting summary has a different risk profile than a donor restriction interpretation. An internal brainstorming document has different consequences than a compliance-related decision. So oversight should vary based on the situation.

A useful approach is to evaluate three factors:

• How easily can an incorrect output be detected?
• What is the potential impact if an error occurs?
• What level of oversight is appropriate based on those factors?

This creates a more practical path forward than applying identical controls to every AI activity.

Vendor AI May Be Your Biggest Exposure

Many institutional leaders assume AI adoption will occur through internally developed solutions. In reality, though, much of today’s AI exposure is arriving through third-party software vendors.

Fundraising platforms, advancement systems, financial applications, productivity tools and analytics products are increasingly embedding AI capabilities into existing services. Sometimes these features appear through software updates before organizations have fully evaluated the implications.

That changes the governance conversation. Questions that once focused on internal projects now extend to vendor relationships:

• How is institutional data being used?
• Is customer data used to train models?
• Can information be shared with clients?
• What controls exist around privacy and confidentiality?
• What audit trails are available?

The institutions making the most progress often involve key stakeholders early. Finance leaders, advancement professionals, IT teams, audit functions, compliance personnel and legal counsel all bring valuable perspectives to the discussion. AI governance becomes more effective when those groups work together before implementation rather than after problems emerge.

The Institutions Moving Fastest Are Not Necessarily the Most Technical

One misconception surrounding AI is that success requires sophisticated internal development teams.

In practice, however, many successful institutions are achieving results through experimentation, focused governance and strong collaboration. They aren’t waiting for perfect data or attempting to solve every challenge at once. They’re identifying targeted use cases, testing carefully, documenting lessons learned and expanding based on results.

They’re also learning from peers. Across higher education, institutions are sharing experiences, governance models, implementation approaches and lessons learned. The pace of adoption is often faster than many leaders realize.

What Comes Next for Foundation Finance Leaders?

The conversation around AI in higher education is changing. The question is no longer whether AI will influence foundation finance and advancement operations, but rather where institutions should begin. It’s important to build the right foundation first.

If your institution is evaluating AI opportunities within finance, advancement, compliance, or reporting functions, now is the time to assess your data readiness, governance approach and operational priorities.

A CPA firm with higher education and AI technology expertise is a valuable partner in this process. A thoughtful approach to AI today can create better outcomes tomorrow.

All content provided in this article is for informational purposes only. Matters discussed in this article are subject to change. For up-to-date information on this subject please contact a James Moore professional. James Moore will not be held responsible for any claim, loss, damage or inconvenience caused as a result of any information within these pages or any information accessed through this site.