Ethical codes, organizational policies and the CRA all discourage commission-based fundraising in Canada. Rogare’s latest report offers a reminder on why this matters. Remunerating fundraisers by a percentage of the money they bring in (“commission”) is almost universally accepted to…
Why do fundraisers leave jobs, and what will motivate them to stay?
Fundraisers anonymously submitted their views and experiences of inappropriate behaviour from donors via an online portal.
Charity values may not be so important to the ethics of turning down donations.
Artificial intelligence (AI) offers exciting opportunities for charities and nonprofits, from automating administrative tasks to gaining insights from data. However, consideration of how it should be used in fundraising has focused mainly on practical applications, with less thought given to the ethical implications that might arise from its use.
When attention does turn to the ethics of using AI for fundraising, the focus is often on how generic ethical issues about AI might also apply to fundraising. But AI will throw up ethical issues and challenges that are unique to fundraising and the nonprofit sector.
A Rogare project team, led by US fundraising consultant Cherian Koshy, has considered what some of these fundraising-specific ethical issues are, and developed a research agenda to explore what else we need to know to rise to these challenges.
This is an excerpt from that report.
Beyond a narrow focus
In December 2022, a Canadian charity, Furniture Bank in Toronto, claimed to have "solved" the poverty porn dilemma using AI. The charity used AI to create images of beneficiaries without photographing real people, arguing this protected beneficiaries' dignity. However, did this really resolve the ethical dilemma? Or did it bring to the surface even more ethical issues arising from AI's entry into fundraising.
This is highly relevant given fundraising's traditional focus on functional skills over normative ethics. Collectively, the fundraising profession has historically been more concerned with the practical aspects of raising money, such as developing effective marketing campaigns, than with the ethical implications of these activities - on what fundraisers can do, but not whether they ought to do it.
With public scrutiny of fundraising ethics growing, AI risks exacerbating problems around over-solicitation, intrusive tactics, and transparency. This is because AI can be used to automate many aspects of fundraising, including identifying potential donors, contacting them, and soliciting donations, with a potential to undermine public trust in fundraising organizations if these are not done appropriately.
Constructing an ethical AI framework is essential to maintain public trust. This framework should include principles such as transparency, accountability, and respect for donor privacy. It should also ensure that AI is used in a way that is consistent with the values of the fundraising organization. By adopting an ethical AI framework, fundraising organizations can help to ensure that AI is used in a way that is beneficial to both donors and the organizations themselves, and ultimately the beneficiaries those organizations help.
The way the fundraising sector has been thinking about the ethics of AI in fundraising is to consider generic concerns about AI and then assume that these apply to fundraising in much the same way that the apply to other sectors.
But, might there be other ethical issues that come from the use of AI in fundraising that have not yet been contemplated, issues that are unique to using AI in fundraising? That's what this project will consider.
This raises an important distinction in how we approach the ethics of AI in fundraising, with two key research questions:
1. What ethical issues are associated with using AI in fundraising?
Beyond concerns such dignity and stereotyping that exist irrespective of AI, there may be novel ethical considerations that emerge specifically from the use of artificial intelligence and automation in fundraising. For example, who owns and controls the data sources powering AI systems? How does the "black box" nature of algorithmic decision-making impact transparency and accountability? Could AI lead to greater exclusion or discrimination through encoded biases? What are the second-order effects on the environment and employment?
2. Can AI be used to resolve ethical dilemmas in fundraising?
Fundraising dilemmas often involve balancing different stakeholder perspectives and rights. But does AI have sufficient understanding of fundraising ethics and normative frameworks to make nuanced ethical judgments? Or could over-reliance on AI for decision-making lead to narrow, technical approaches lacking human wisdom? More research is needed on the limits of AI in contextualizing principles and values when navigating tricky real-world scenarios.
Investigating these two questions will be essential as the fundraising profession adopts new AI capabilities. Doing so from the outset, rather than retrofitting solutions, will help maintain public trust and focus innovation on serving social good.
The project team's findings and recommendations can be found in the full report published in February 2024 - Artificial intelligence and fundraising ethics: A research agenda.
To talk about the research agenda and this project, contact Cherian Koshy at cheriangkoshy@gmail.com. For further information about Rogare, contact director Ian MacQuillin at ianmacquillin@rogare.net.
Jennifer Johnstone, the chair of the Association of Fundraising Professionals (AFP) Canada, released the following statement in response to articles published by the Canadian Press, the Globe and Mail, La Presse, Le Devoir and CTV News:
AFP Canada appreciates media stories on charities as they help to raise awareness of our sector's contribution to our country. We were, however, disappointed in the recent articles that focused on the compensation packages of the heads of some Canadian environmental charities. These articles failed to consider important factors, most notably the impact of the work of these organizations.
Charities are tasked with addressing some of the biggest problems we face, in this instance the climate crisis among other critical environmental issues. These are extremely complex and challenging issues that threaten our planet and our very existence.
It is in the best interest of all Canadians that our country's most talented leaders address these issues. Sophisticated organizations are necessary to address complex challenges. We need people who are more than simply "... the best in terms of mobilization, faith and belief in adherence to the mission...". Strong leadership skills and expertise in finance, administration, governance, human resources and fundraising are crucial to these organizations' success in addressing their missions and making changes to improve our communities.
Canadian charities have an imperative to be able to sustainably fund their work. To affect change they need professional leaders and staff who are remunerated fairly and equitably for their contributions. Canadians from coast to coast to coast depend on it.
Instead of looking at salaries it would be far more beneficial to focus on the impact of these organizations. Is their work affecting change? How has their work made an impact on the environment? Focusing on impact would make for a much more newsworthy story.
The Association of Fundraising Professionals empowers individuals and organizations to practice ethical fundraising through professional education, networking, research and advocacy.
In the report "Artificial Intelligence and Fundraising Ethics," Rogare's project team identified potential ethical issues relating to the use of AI in fundraising in seven areas: data and data ethics, equity, transparency, accountability, identity disclosure, public trust, second-order effects. Rather than describe generic ethical issues regarding AI (all use of AI in all sectors raises issues of data ethics and human oversight) they tried to relate them to the specifics of using AI in fundraising.
In this excerpt, the impact of AI on equity is considered.
Equity
Access - Costs of AI could concentrate capabilities among large, well-funded nonprofits, excluding smaller organisations from AI benefits. Yet equal access enables AI to help overcome barriers such as language differences between donors and beneficiaries. Policies must balance open access with necessary funding of AI development.
Widening gaps - The funding gap between large and small nonprofits could grow as bigger budgets allow quicker AI adoption. But shared data infrastructure and open standards could mitigate this.
Manipulation - AI could enable more effective emotional manipulation, disproportionately targeting disadvantaged or vulnerable populations. Safeguards against exploitative communication must be robust.
Exclusion - AI risks sidelining communities from telling their own stories if used for automated personalised messaging. The dignity of beneficiaries could be violated by not enabling direct participation. Oversight is critical.
Encoded biases - Historical biases around race, gender, income and other attributes in fundraising data could lead AI to reinforce discrimination and widen inclusion gaps. Detecting and correcting these biases is not a one-time task but an ongoing process that requires diverse oversight and continual refinement.
While acknowledging these challenges, it's also important to highlight the potential of AI to foster equity and inclusivity in fundraising.
AI, implemented properly, could empower smaller charities, enabling them to produce more effective fundraising content swiftly and cost-effectively, thereby levelling the playing field. Moreover, AI tools offer opportunities for under-represented groups in fundraising, such as those not traditionally university-educated, to enhance their skills and contribute valuable perspectives.
This approach is not about compensating for perceived deficiencies; rather, it's about leveraging technology to amplify diverse voices and enable all fundraisers to reach their full potential. AI should be seen as a tool to bridge gaps, not to plaster over them. It is vital to avoid over-generalisations or assumptions about groups based on education or background. By harnessing AI responsibly and ethically, we can make fundraising more inclusive, representing a wider range of experiences and backgrounds.
Training and educational initiatives should accompany the deployment of any AI tools, providing fundraisers with the knowledge and skills to use these technologies and involve lived experiences effectively. This holistic approach ensures that AI is not only a technological solution but also a part of a broader strategy to enhance the fundraising profession for everyone, irrespective of their educational or socio- economic background.
This paper, "Artificial Intelligence and Fundraising Ethics," along with all of Rogare's reports, research and other outputs, is available free of charge to the fundraising profession. Rogare thinks it is important that people should be able to access all the ideas coming out of Rogare, and are able to provide this access through the ongoing generous support of—Ask Direct (Ireland), Bluefrog Fundraising (UK), ST (Stephen Thomas Ltd) (Canada), GoalBusters (USA) and Giving Architects (NZ). Details of all their projects can be found on the Rogare website - www.rogare.net