National AI Plan community organisations Australia – what not-for-profits need to know
Australia’s National AI Plan names community organisations and
not-for-profits in its Spread the Benefits pillar, providing
support through the AI Adopt Program and National AI Centre.
But the National AI Plan for community organisations arrives
alongside a Privacy Act compliance deadline that most NFPs are
not yet prepared for. From 10 December 2026, the Privacy and
Other Legislation Amendment Act 2024 requires organisations
using automated decision-making that affects client rights to
disclose this in their privacy policy. With only 14% of
Australian not-for-profits having an internal AI governance
framework, the governance gap is becoming a compliance gap for
community organisations and NGOs across Australia.
What We Are Finding
What Australia’s National AI Plan means – and does not mean – for community organisations.
March 2026
Policy response: Australia’s National AI Plan, December 2025
The Australian Government named not-for-profits in its National AI Plan. That is good news. But the same plan set a compliance clock running that most community organisations do not yet know about. Here is what the National AI Plan means for community organisations across Australia, what it does not say, and what you need to do before December 2026.
Department of Industry, Science and Resources, National AI Plan 2025, Australian Government, December 2025.
Privacy Act automated decision-making obligations December 2026 community organisations
The government named you. That matters.
On 2 December 2025, the Australian Government released the National AI Plan – its most comprehensive statement to date on how it intends to manage the expansion of AI across the economy. The plan is organised around three pillars: capture the opportunity, spread the benefits, and keep Australians safe.
For the for-purpose sector, the most significant thing about the plan is not what it mandates. It is what it acknowledges. The plan explicitly names not-for-profits in the “Spread the Benefits” pillar. It confirms that the AI Adopt Program and the National AI Centre will provide tailored support for NFP adoption. It recognises that digital exclusion and uneven AI adoption across regions and communities is a persistent problem that requires targeted action.
After years of policy that treated community organisations as an afterthought in technology transitions, being named is meaningful. It signals that the government understands the for-purpose sector is a distinct environment with distinct needs.
The plan explicitly names not-for-profits. After years of being an afterthought in technology policy, that recognition matters. But recognition is not the same as resourcing.
The harder question is whether recognition translates into real support at the scale the sector needs. And the honest answer, at this point, is: not yet.
What the plan says
Capture the opportunity
Investment in compute infrastructure, data centres, connectivity and domestic AI capability. Australia to become a regional AI hub.
Spread the benefits
Support for SME and NFP adoption via the AI Adopt Program and National AI Centre. Workforce skills, AI literacy, and reskilling pathways. Explicitly addresses digital exclusion and uneven regional adoption.
Keep Australians safe
No standalone AI Act. Reliance on existing laws supplemented by the new AI Safety Institute ($29.9M). Every organisation using AI is responsible for identifying and responding to AI harms.
Source: National AI Plan 2025, Australian Government
AI governance gap Australian not-for-profits National AI Plan compliance
Australia’s National AI Plan names community organisations and
not-for-profits in its Spread the Benefits pillar, providing
support through the AI Adopt Program and National AI Centre.
But the National AI Plan for community organisations arrives
alongside a Privacy Act compliance deadline that most NFPs are
not yet prepared for. From 10 December 2026, the Privacy and
Other Legislation Amendment Act 2024 requires organisations
using automated decision-making that affects client rights to
disclose this in their privacy policy. With only 14% of
Australian not-for-profits having an internal AI governance
framework, the governance gap is becoming a compliance gap for
community organisations and NGOs across Australia.
What does not mean – and what does not wait.
The plan does not create new legal obligations directly. Australia has no standalone AI Act. The government’s position is to rely on existing laws and issue guidance – a deliberately light-touch approach designed to avoid stifling adoption.
But something else has been running in parallel that does create new obligations – and the clock is ticking regardless of whether your organisation received any support from the AI Adopt Program.
From 10 December 2026, amendments to the Privacy Act 1988 come into force. Under new APP 1.7, if your organisation uses AI or any computer program to make decisions that could reasonably be expected to significantly affect the rights or interests of an individual – using personal information – you must disclose this in your privacy policy. The obligation applies to decisions made on or after that date, regardless of when your systems were set up.
What is in scope for community organisations
The obligation applies where a computer program is used to make or substantially assist in making a decision that could significantly affect an individual’s rights or interests. The Privacy Act specifically names examples including decisions that affect access to a significant service or support.
For community organisations this could include: AI-generated case note summaries that inform eligibility decisions. CRM tools that score or prioritise clients for follow-up. Chatbots that direct people to or away from services. Risk assessment tools used in intake processes.
Note: organisations with annual turnover under $3M may be exempt unless they are health service providers or handle sensitive client data – which many community organisations do.
Non-compliance exposes organisations to fines, compliance notices, and infringement notices from the OAIC. The regulator began its first ever privacy compliance sweep in January 2026. The signal is clear: privacy transparency is now an enforcement priority.
The governance gap
of Australian NFPs have an internal AI policy or governance framework
Despite 76% now using AI tools – a 52% increase in adoption in a single year. The plan raises expectations for governance without closing the gap.
Infoxchange, Digital Technology in the Not-for-Profit Sector Report, 2025
2026 – Privacy Act automated decision-making obligations commence
The obligation applies to decisions made from this date regardless of when the underlying system was deployed. Less than nine months away.
Privacy and Other Legislation Amendment Act 2024 (Cth)
of Australians believe current AI laws and safeguards are adequate
78% of Australians are concerned about negative outcomes from AI. Only 36% trust AI systems broadly. Community organisations are navigating adoption against a backdrop of deep public scepticism.
University of Melbourne and KPMG, 2025
Not For Humans examines the AI labour market impacts identified by Anthropic researchers in March 2026. The findings show AI labour market community organisations need to understand: displaced workers skew female, educated, and mid-career. This is a description of the for-purpose sector workforce. The AI labour market pipeline problem means entry-level contraction is already measurable.
What this actually means for your organisation.
The National AI Plan is not a compliance document. It does not create obligations for community organisations directly. What it does is signal very clearly where the policy environment is heading: higher expectations for responsible governance, more scrutiny from regulators, and an acknowledgement that the for-purpose sector needs support to get there.
That support is real but limited. Understanding what the National AI Plan means for community organisations across Australia requires separating what the plan promises from what it delivers. The AI Adopt Program and the National AI Centre are genuine resources. But they are oriented toward AI adoption – helping organisations use AI – rather than toward the governance of AI already in use. Most community organisations are not at the “how do we adopt AI” stage. They are at the “AI arrived in our tools before we noticed and now we are not sure what we have or what it is doing” stage.
The Privacy Act deadline is the clearest near-term obligation. But it is not the only signal. The plan is explicit that AI adoption must be consultative – workers and unions should be involved in decisions where AI affects job roles, processes, or conditions. Every organisation using AI is responsible for identifying and responding to AI harms. The expectation of human oversight over AI-assisted decisions affecting individuals is built into both the plan and the incoming privacy framework.
These are not technical requirements. They are governance and behaviour requirements. They are exactly the layer where community organisations are most under-resourced and where the Not For Humans framework is designed to help.
Three things to do before December 2026
Audit what AI tools you are already using
Map every tool across your organisation. Include CRMs, case management systems, chatbots, scheduling tools, and document generation. Identify which ones make or inform decisions about clients.
Check whether your privacy policy covers automated decisions
If your organisation uses AI tools that could affect client access to services, your privacy policy will need to describe how personal information is used in those decisions before 10 December 2026.
Bring your board into the conversation now
The Australian Institute of Company Directors confirmed in 2024 that boards have a duty of care that extends to how AI is used within their organisations. The plan reinforces this. Governance failure is not just a compliance risk – for community organisations, it is a trust risk.
Not For Humans
The for-purpose sector needs governance tools that work without specialist expertise.
The National AI Plan names community organisations. The Privacy Act sets a deadline. The Not For Humans Toolkit gives you a practical, self-guided way to build governance at the layer where it actually matters – human behaviour and decision-making – before the obligations arrive.
Not For Humans: Because AI’s greatest risk isn’t technical, it’s Human.
