AI Governance for Australian Non-Profits and Charities: The AI6 Framework, ACNC Obligations, and Practical Implementation

Australian non-profits and charities are adopting AI at pace. AI is being deployed across donor management (predictive donor scoring, automated stewardship communications), service delivery (chatbots for client intake, AI-assisted case management), fundraising (campaign optimisation, donor segmentation), operations (financial automation, grant application drafting), and program design (data analysis for impact measurement). The sector faces a specific governance challenge: limited compliance resources, sensitive beneficiary populations, donor trust as a critical asset, and regulatory frameworks that apply differently to charitable purposes than to commercial activity. The Australian Government's National AI Centre released Guidance for AI Adoption (AI6) in October 2025 with specific resources for the social sector. This guide explains how to apply it.

The AI6 framework for non-profits

The AI6 framework consists of six essential practices: accountability, governance, risk management, transparency, data governance, and human oversight. For non-profits, each practice has specific implications. Accountability: charities operating under ACNC Governance Standard 1 must have written governing documents — these should now address AI use. Designate a board member or senior executive accountable for AI governance. Governance: integrate AI oversight into existing board governance — most non-profit boards already have risk, finance, and audit committees that can extend their remit. Risk management: AI risks for non-profits include misallocating resources based on AI-influenced decisions, harming beneficiaries through biased AI service delivery, and breaching donor expectations through unauthorised AI use of donor data. Transparency: ACNC requires charities to be transparent about how they pursue their charitable purposes — this now extends to disclosing material AI use to donors and beneficiaries. Data governance: charities hold sensitive beneficiary data (often including vulnerable people), donor data (with associated commitments about use), and volunteer data — each category needs AI governance attention. Human oversight: AI cannot replace the personal relationships and human judgement that are core to most non-profit missions.

ACNC Governance Standards and AI

The five ACNC Governance Standards apply directly to AI governance for registered charities. Standard 1 (purposes and not-for-profit nature): AI use must align with the charity's purposes and cannot pursue commercial benefit at the expense of charitable mission. Standard 2 (accountability to members): material AI use should be disclosed in annual reporting and through member communications. Standard 3 (compliance with Australian laws): includes Privacy Act, Australian Consumer Law, anti-discrimination law, and emerging AI-specific obligations. Standard 4 (suitability of responsible persons): responsible persons must understand the AI risks the charity is managing. Standard 5 (duty of care and diligence): the board must exercise reasonable care and diligence in AI deployment decisions — including understanding the risks before approving AI use cases.

Privacy Act ADM transparency — effective 10 December 2026

The Australian Privacy Act's automated decision-making transparency obligation takes effect 10 December 2026. Charities that make decisions substantially based on automated processing (about beneficiaries, donors, employees, or service users) must notify affected individuals that automated processing was involved. This applies regardless of charity size if the organisation is covered by the Privacy Act (broadly, those with annual turnover over $3M, plus health service providers and certain other categories). Smaller charities not directly covered should still consider this as best practice, particularly when handling sensitive beneficiary information.

Practical AI use cases and governance

Donor management AI: predictive donor scoring and automated stewardship communications can be highly effective but raise governance questions about consent for data use, potential exclusion of donors based on algorithmic scoring, and transparency about how AI influences donor relationships. Beneficiary services: AI-assisted client intake, chatbots for initial contact, and automated needs assessment can extend the reach of limited resources but must be designed to avoid harming vulnerable populations, exclude individuals who cannot effectively use digital channels, or replace human contact where human contact is essential. Fundraising AI: campaign optimisation and donor segmentation should respect donor preferences and avoid manipulative patterns. Operations AI: financial automation and grant application drafting (an increasingly common use case) raise quality control and disclosure questions — many grant funders expect human authorship and may require disclosure of AI assistance.

Resources from the National AI Centre

The National AI Centre provides specific resources for non-profits as part of the AI6 framework, including an AI screening tool (to identify whether AI is suitable for a particular use case), an AI policy template (customisable starting point for charity AI policies), an AI system register template (for maintaining inventory of AI in use), and a glossary of terms. These are available at industry.gov.au/NAIC. The National AI Plan (December 2025) consolidated SME and not-for-profit support within the National AI Centre and includes extended First Nations support initiatives.

Primary sources: National AI Centre — Guidance for AI Adoption | ACNC Governance Standards

Related reading

• Australia AI Safety Standard
• APRA and ASIC: AI for Australian FS
• Board AI Literacy for Directors