Case study · Not-for-profit
Giving charity CEOs the confidence to use AI properly
Charity chief executives were being asked to make decisions about AI without the grounding to make them safely. We ran capability workshops with CEOs across the Canberra community sector on where the ethical and legal lines sit, and where AI can improve the services they deliver.
- Client
- Hands Across Canberra
- Sector
- Not-for-profit
- Published
- 8 July 2026
- CEO-level
- We aimed it at the people making the decision
- Ethical & legal
- We grounded it in their actual obligations
- Sector-wide
- We delivered it across organisations at once
Context
Hands Across Canberra works across the ACT community sector, and the question kept arriving from the organisations it supports: what are we allowed to do with AI?
It was a fair question with no easy source of an answer. Charity chief executives were watching their staff adopt AI tools informally, watching peer organisations do the same, and being asked by boards whether the organisation had a position. Very few had access to the expertise that would normally answer this. Most community organisations have no in-house counsel, no privacy officer and no technology executive.
That leaves a chief executive choosing between two poor options. Prohibit AI, and forgo a genuine opportunity to increase service capacity without increasing headcount, which matters in a sector where demand rises faster than funding. Or allow it informally, and carry a risk nobody has assessed, involving some of the most sensitive personal information there is, about people least able to bear the consequences of a mistake.
Our approach
Aimed at the decision-makers
We ran the workshops for chief executives rather than for staff. This was deliberate. Staff training tells people how to use a tool; it does not resolve whether the organisation should be using it, on what data, under what conditions. That decision sits with the chief executive and the board, and it cannot be delegated to whoever is most enthusiastic about the technology.
Built on obligations, not on AI generally
The sessions were not general AI awareness. They were grounded in the specific things a community organisation has to satisfy:
- Privacy, and the practical difference between de-identified and identifiable client information once it enters a general-purpose AI tool.
- Duty of care to clients who may be vulnerable, and what that implies about using AI anywhere near a decision affecting a person's support.
- Funder and contractual obligations, which frequently constrain how information may be handled regardless of what privacy law permits.
- Consent and transparency, and whether clients would reasonably expect their information to be handled this way, which is often a sharper test than legality.
Where AI genuinely helps a charity
The other half was the opportunity, framed honestly. The administrative load in the community sector is substantial and it directly displaces service delivery: grant applications and acquittals, funding reports, board papers, policy documents, meeting notes, and searching an organisation's own accumulated records.
That work is largely internal, uses information the organisation already holds, and does not touch a decision about a client. It is where the benefit is real and the risk is manageable, and it is a very different proposition from anything touching case management.
The most useful thing we can give a chief executive is not a policy. It is enough understanding to recognise which side of the line a given idea falls on, and the confidence to say so.
Practical, not theoretical
Sessions were built around the questions leaders actually brought: real proposals their staff had raised, real tools already in use. Working through concrete cases rather than abstract principles is what turns a session into a decision people can defend afterwards.
We also ran this across organisations rather than one at a time, which shares the cost and lets chief executives compare notes with peers facing identical questions. In a sector where nobody can afford a consulting engagement of their own, that model matters.
Outcomes
Chief executives across the Canberra community sector came away able to tell which side of the line a given idea falls on, and to say so with confidence to a board or a funder.
That is the outcome we want. A CEO who understands where the boundaries sit makes dozens of sound decisions over the following year without needing us again.
More work
Building AI capability at two scales: the enterprise and the community
Policy does not make people AI-capable. We pair whole-of-agency sessions that reach everyone with a small, open Canberra practitioner community co-founded with an Australian university, because confidence is built by trying things rather than by reading guidance.
- One agency heard the same message at once
- Whole-of-staffOne agency heard the same message at once
- The partnership keeps sessions evidence-based
- UniversityThe partnership keeps sessions evidence-based
- AI in the Shadows meets regularly in Canberra
- ~30 peopleAI in the Shadows meets regularly in Canberra
Cold chain monitoring, reporting and alerting, built at speed
A national courier business needed visibility over temperature-controlled freight: continuous monitoring, reporting its customers could rely on, and alerting that reached someone while the load could still be saved. We used AI-accelerated development practices to build it in a fraction of the usual timeline.
- We went from start to a working system in weeks
- WeeksWe went from start to a working system in weeks
- Monitoring runs across temperature-controlled freight
- ContinuousMonitoring runs across temperature-controlled freight
- Alerts arrive while a load can still be saved
- Real-timeAlerts arrive while a load can still be saved
Contact
Tell us what you are trying to do and we will tell you whether AI helps.
If AI is the wrong tool for your problem we will say so, and we will explain why.