AI, Scale, Trust and Risk: The Nonprofit Imperative for Responsible and Beneficial AI

This past week in San Francisco was a convergence of two critical conversations around the future of trustworthy AI. From hosting the 2nd Annual Fundraising.AI Global Summit—which brought together nearly 10,000 nonprofit professionals and technologists from over 100 countries—to attending Credo AI 's 3rd Annual Responsible AI Leadership Summit, I found myself both inspired and concerned. Inspired by the momentum behind Responsible AI (RAI) and concerned about the narrow focus on that seems to define much of the current discourse.

While the private sector is deeply invested in promoting trustworthy AI in the short term—focusing on immediate risks, regulatory requirements, and shareholder returns—the nonprofit sector has a far broader responsibility. We must not only adopt Responsible AI practices but also consider AI’s long-term societal impact and unintended consequences. This is where Beneficial AI comes into play—AI that serves humanity for the greater good, not just the interests of those with the resources to deploy it.

"The subtle yet profound difference between "responsible or trustworthy AI" as a stand-alone ideal, and "Responsible and Beneficial AI" is like comparing compliance to conscience."

Trustworthy AI ensures that systems are safe, transparent, and aligned with regulations. But Beneficial AI goes beyond safety—it actively seeks to improve societal well-being, ensuring that technology not only avoids harm but creates meaningful, positive change. This shift from merely responsible to truly beneficial AI is critical for addressing the long-term, systemic challenges that nonprofits face in their missions to serve humanity.

As an observer of Responsible AI efforts on both sides, I’m concerned that the private sector—those largely building the systems that are democratized quickly and broadly—carries little, if any, true burden for the future welfare of humanity.

AI at Scale: Short-Term Control vs. Long-Term Responsibility

AI is advancing at a pace that far exceeds our ability to fully grasp its implications. While at AI for Good in Geneva last June, we heard that for every $1,000 spent on scaling AI, only $1 is invested in AI safety. In a compelling presentation on the same topic, Tristan Harris from the Center for Humane Technology shared, "There is a significant gap between those who are working on AI capabilities and those working on safety. For every 30 papers published on AI capabilities, only one paper is published on safety. We need to rebalance this, as the stakes are incredibly high."

This imbalance is a symptom of a broader issue—the for-profit world is overwhelmingly fixated on controlling AI’s immediate risks to meet regulatory standards and protect their bottom lines. They are, understandably, preoccupied with short-term control. But what happens beyond that?

"The nonprofit sector, free from the relentless pressure of quarterly earnings and competition, is uniquely positioned to ask the harder, long-term questions. Questions like: What are the unintended societal consequences of deploying AI at scale? How will AI affect vulnerable populations over the next decade? And how can we ensure that AI’s benefits are shared equitably across all communities, not just those with the resources to harness its power?"

As I shared in my previous article, "Beyond AI Ethics: The Nonprofit Sector's Imperative ," the work of Responsible AI cannot stop at ethics and governance—it must extend to ensuring AI’s long-term benefit to humanity. AI that prioritizes not just safety, but also fairness, equity, and the greater good. This is the burden—and the opportunity—of the nonprofit sector.

Why the Nonprofit Sector is Uniquely Positioned to Lead

The lack of financial incentives in the nonprofit world gives us a unique advantage in leading the Responsible and Beneficial AI movement. Without the pressures of delivering shareholder value, nonprofits can focus on what truly matters—ensuring that AI is deployed not just responsibly, but in ways that actively benefit society in the long term.

At the Credo AI Responsible AI Leadership Summit, Christina Montgomery , IBM ’s Chief Privacy & Trust Officer, spoke about the need for a "bottom-up approach" to creating a culture of trustworthy AI, with leadership and vision coming from the top. This is where nonprofits excel. As mission-driven organizations, we are already aligned to prioritize ethics, safety, and the public good over rapid growth and profitability. Our vision naturally extends beyond the immediate future, into the realm of long-term societal impact.

As Paula Goldman , Chief Ethical and Humane Use Officer at Salesforce , put it, “Questions of trust are business critical.” While this statement speaks to the for-profit sector, it rings even more true for nonprofits, where trust is not just business critical, but mission-critical. Our ability to do good depends on maintaining the trust of the communities we serve, the donors who support us, and the public who rely on us to be ethical stewards of change.

Moving Beyond Responsible AI: Toward Responsible and Beneficial AI

While the conversation around Responsible AI is vital, it is only the first step. Responsible AI focuses on safety, governance, and mitigating harm—ensuring AI does no damage. But Beneficial AI goes further. It asks: How can AI be a force for good? How can we leverage AI to reduce inequality, increase access to opportunities, and solve some of society’s most pressing challenges?

At Fundraising.AI , we exist to bridge that gap. Our grassroots movement has brought together nonprofit professionals, technologists, and thought leaders from over 100 countries to discuss how AI can be used both responsibly and beneficially in the nonprofit sector.

This global community, driven by mission rather than profit, is leading the charge to ensure that AI amplifies our ability to create positive, equitable change in the world.

Yet, as I observed at the Credo AI summit, there is a striking imbalance between those developing AI and those concerned with its governance and long-term consequences. The majority of developers are not involved in these critical conversations around AI safety and societal impact—something as obvious as their attire in the room underscored this divide.

"There’s an urgent need for more voices, especially from the nonprofit and social sectors, to step into these discussions and shape the future of AI."

The Nonprofit Sector’s Responsibility to the Future

Nonprofits have always been society’s moral compass, advocating for vulnerable communities, championing social justice, and pushing for equity. Now, in the age of AI, we must continue to play that role, ensuring that AI serves humanity rather than exacerbating existing inequalities.

This is not a conversation for the future—it’s a responsibility we must shoulder now. If we wait until AI’s unintended consequences are upon us, we will have missed our opportunity to shape its trajectory. The speed and scale of AI advancement demand that nonprofits step up to lead—not only in how AI is adopted but in how it is governed and deployed to ensure long-term benefit.

"The nonprofit sector must embrace this moment with urgency. We are not just another stakeholder in the AI conversation—we are the sector best equipped to lead it."

The private sector may have the resources to build and control AI in the short term, but it is the nonprofit sector that must think beyond today’s risks and prioritize AI’s impact on future generations. This is the challenge and the opportunity before us.

As we continue building on the work of Fundraising.AI , I encourage nonprofit leaders, technologists, and funders to prioritize both Responsible and Beneficial AI. Together, we can shape an AI-driven world where technology serves the greater good, and where progress is measured not by the speed of deployment but by the positive impact it leaves in its wake.