AI for nonprofits and community organisations

What it can do

The documented picture

A 2026 benchmark study of 346 nonprofits found that 92% now use AI in some capacity, while only 7% report major operational impact and 47% have no AI governance policy at all. 81% describe their use as ad hoc, with no documentation of what works.

Read that carefully, because it inverts the usual pitch. Access is not the problem. Nearly everyone has access. The problem is that nobody owns the work, nothing is written down, and there is no policy telling staff what is allowed.

What it can do

What actually moves for a nonprofit

• Reporting and documentation burden — the single most common real win
• Turning programme notes into funder-ready narrative
• Reading and summarising long RFPs, regulations and funder guidance
• Drafting communications across the channels you already use
• Making institutional knowledge searchable so it survives turnover

What it cannot do

What it will not touch

Underfunding. Low wages. Philanthropic inequity. Burnout that is really understaffing. Any proposal implying otherwise should be treated as a signal about the vendor, not about the technology.

What you carry

Governance first, tooling second

Write the policy before you expand the tooling. Name who is accountable. Decide what may never be entered — donor records, client information, case notes. Tell the people you serve where this is used on their information, and give them a route to decline that does not cost them the service.

An organisation that does those four things with modest tools will outperform one with better tools and none of them.