Harmonizing AI and Philanthropy: Running To or Running From a Brave New World

In the ever-evolving landscape of technological advancement, Artificial Intelligence (AI) stands out as a beacon of innovation, promising transformative changes across industries and societies. Concurrently, philanthropy continues to serve as a cornerstone of human compassion, striving to address pressing social issues and create sustainable impact. It is at the intersection of these two (domains) realms where unprecedented opportunities exist for philanthropy to use this technology for the betterment of society and the nonprofit organizations they fund. Yet, AI is largely absent from discussion and action within the philanthropic sector either because it viewed with suspicion, misunderstood, or unintelligible to most people.

AI, however, is the most pressing racial equity issue of our time due to its ubiquitous nature made possible by the convergence of technological innovation, data abundance, affordability, and market demand making it an integral part of our society and driving its continued growth and adoption across diverse domains. More importantly, the data algorithms upon which AI relies inherit biases present in its underlying data sources as researchers Dr. Joy Buolamwini (Unmasking AI My Mission to Protect What Is Human in a World of Machines) and Dr. Safiya Noble (Algorithms of Oppression) have documented. These algorithms are used to inform decisions from who votes, who gets to own a house, what school a person can attend, who gets healthcare and the quality therein, which community gets access to a grocery store, who gets a job and more. It cannot be ignored.

So why should philanthropy care? Because AI systems can perpetuate and even exacerbate existing biases and discrimination present in the data upon which they are modeled. If historical data reflect systemic inequality and bias, AI algorithms learn and perpetuate these patterns, leading to discriminatory outcomes that disproportionately and negatively affect oppressed and under-represented communities and people of color—communities about which philanthropy cares.

Philanthropy can be a powerful disruptor in the current ecosystem of AI development to be ethical, inclusive, and beneficial to the philanthropic sector and the nonprofits they fund. Addressing the risks inherent in AI will require consideration of ethical principles, robust oversight mechanisms, and ongoing dialogue between technologists, philanthropists, policy makers, and affected communities to ensure that AI technologies are deployed in ways that prioritize fairness, transparency, and human well-being. In its role of convenor, philanthropy can certainly call these players together.

Collectively, philanthropy has the resources to fund research, engage diverse community stakeholders, and advocate for governance structures and policies that force AI development to be transparent and accountable. For this to happen, philanthropy must step out of its comfort zone and dare to push for change beyond its own interests.

We can begin to understand how AI is used as a tool to analyze data and ask questions about the biases in its own data. Additionally, it can analyze grantee reports in ways that tell new narratives and identify what really changed because of its funding. At my previous foundation, Episcopal Health Foundation, I launched an AI project that analyzed six years of grantee reports that helped the organization understand if and what change had occurred based on its funding strategies. In so doing, it provided the opportunity to lift the burden on program officers to do this work manually and allowed them to use AI to supplement and expedite the learning process. It also helped the foundation understand and articulate to board members and the community how its funds were deployed and whether the transformational change they sought had occurred.

Beyond that, philanthropy can fund AI research that explores new avenues to develop inclusive algorithms to bring new insights and solve complex problems. It can support initiatives that address bias, fairness, transparency, and accountability in AI systems. It can begin its own learning journey and build capacity by funding infrastructure development within its foundation and for its nonprofit partners. It can use its voice to advocate for equitable policies, regulations, governance structures, AI safety and security. It can rise to this moment by taking bold action to change the way things are and imagine and foster a different more human-centered AI ecosystem that benefits all of society.

There is a growing interest in AI on the part of philanthropy as demonstrated by the crowded tables and standing room only sessions at the 2024 GEO National Conference that addressed this topic. The challenge we face now is imagining a brave new world where AI is developed and used in ways that are ethical, inclusive, and work for society—not marginalize it—while also addressing the risks and challenges associated with these technologies. Not to do so is a missed opportunity with long-lasting implications.

–

Jo Z. Carcedo

Former Vice President for Grants

With more than three decades of senior executive leadership in non-governmental organizations, Jo has been a catalyst for change, recognized for designing programs and processes that help organizations meet their strategic priorities and chart new paths. Her unique multi-disciplinary background spans healthcare, human services, and philanthropy. Her philanthropic experience includes the creation and oversight of the grantmaking system for a leading Texas philanthropy, Episcopal Health Foundation (EHF), where she guided the selection and distribution of $500 million in grant awards. She is highly respected for the innovative practices that she brought to philanthropy. She has published in The Foundation Review, blogged for Health Affairs, and presented at various state and national conferences including the National Academies of Sciences, Engineering, and Medicine, PEAK Grantmaking, Texas Primary Care Consortium, among others.