What does Effective Altruism reveal about the future of philanthropic and technological governance?

Effective Altruism has reshaped global philanthropy and AI policy but faces criticism over governance gaps and unintended consequences.

On Nov. 9, 2025, the Trump administration flagged AI company Anthropic and its ties to Effective Altruism, citing concerns about ideological influence on federal rule-making. "Anthropic is trying to smuggle ideology into federal rule-making under the banner of 'safety,'" a former White House official told the New York Post. The critique wasn't an outlier—Republican skepticism toward the effective altruist philosophy had been growing, driven by questions about its governance structures, opaque decision-making, and extraordinary financial concentrations among Democratic-aligned mega-donors.

Effective altruism started as a niche intellectual movement but has become one of the most influential frameworks shaping philanthropy, AI governance, and long-term policy planning. The movement’s insistence on "doing the most good" derives from utilitarian ethics, as advanced by philosophers like Peter Singer and institutionalized through organizations led by Will MacAskill at Oxford University. Advocates hold that individuals and institutions should prioritize interventions with the highest measurable impact per dollar or effort—a principle that initially drove funding toward global health programs but later shifted to existential risks such as artificial intelligence.

Anthropic, founded in 2021 by Dario Amodei and seven former colleagues from OpenAI, was born out of frustration with commercially driven AI scaling. Its founders believed accelerated AI development endangered safety. In 2022, the company secured $580 million in Series B funding, largely from Sam Bankman-Fried’s FTX Future Fund—a catastrophic connection for the movement when Bankman-Fried’s alleged misuse of customer funds destroyed his credibility.

The FTX fallout led to widespread scrutiny of Effective Altruism’s governance and “earning to give” philosophy, as critics argued that its ends-justify-the-means logic contributed to moral hazards. As Vox later summarized Bankman-Fried’s implosion: "How Effective Altruism let Sam Bankman-Fried happen."

While Anthropic distanced itself from the Effective Altruism brand following the collapse, its institutional DNA remains intertwined with EA principles. Co-founders Amodei and President Daniela Amodei both have documented ties to EA institutions; Open Philanthropy co-founder Holden Karnofsky—a close proxy for the movement—is married to Daniela and joined Anthropic’s staff in 2025. The company, which pledged 80% of co-founder wealth to philanthropy, explicitly labels itself a Public Benefit Corporation tasked with scalable, long-term benefits to humanity—a mission rooted in EA’s core tenets.

Yet the overlap of EA and Anthropic spurred criticism on ideological governance. Jaan Tallinn, an Anthropic investor and EA-aligned philanthropist, co-founded the Future of Life Institute, which calls for pauses on frontier AI systems. Still, Tallinn has simultaneously bankrolled leading AI accelerators, including Anthropic itself, deepening what critics describe as EA’s inherent contradiction: promoting risk mitigation while funding its escalation.

From a philosophical perspective, EA’s focus on existential dangers has drawn fire for prioritizing speculative risks over real, immediate suffering. Timnit Gebru and Émile Torres argue what they term "TESCREALism"—an ideological cluster tied to EA—systematically deprioritizes current societal harms like systemic injustice or labor exploitation. In contrast, longtermist campaigns, such as Anthropic’s public safety commitments, focus predominantly on theoretical "future harms."

Dario Amodei defended this logic in 2026, telling Fortune, “Most people underestimate both the radical upside of AI and its dangerous risks.”

Meanwhile, the movement’s institutional concentration has raised concerns. Open Philanthropy, which directed over $4 billion in grants, remains heavily funded by Dustin Moskovitz—Facebook’s co-founder and one of EA’s anchor donors. Alongside Moskovitz and Bankman-Fried’s mega-contributions, organizations like Georgetown CSET or Anthropic operate on grant pathways so consolidated they’ve reshaped global AI policy debates.

Aggressive federal lobbying only compounds the critique. Anthropic alone escalated spending to $3.13 million in 2025, prompting David Sacks, Trump’s AI czar, to warn that EA “weaponizes safety rhetoric for regulation capture."

The broader critique argues EA inadvertently reproduces technocratic elitism rather than delivering substantive accountability. The Long-Term Benefit Trust—a governance framework at Anthropic—faces its own credibility gaps. Although Trust members hold equity safeguards, much of Anthropic’s board influence remains underutilized.

An internal EA Forum review detailed underwhelming trustee appointments and flagged discrepancies between board independence promises and corporate investor alignment.

"Who elected the longtermists?" asked David Thorstad, philosopher at Vanderbilt University. The democratic absence in EA-infused corporate missions like Anthropic’s remains a recurring friction, especially given the world-altering implications of scaled AI integration.

The Pentagon’s recent designation of Anthropic as a "supply chain risk" underscores fears that institutional configurations reliant on EA logic leave critical policy concentrated among a small and ideologically coherent elite.

As Effective Altruism projects grow—with Anthropic collecting hundreds of millions annually from chatbots like Claude and Mythos—the tension between accountability and philosophical purity deepens. Critics like Gary Marcus frame Anthropic’s walkback of specific safety commitments as a caution.

“The connection exists—fundraising overlaps regulate theory,” Marcus wrote. “Anthropic sells the safety utopia, then scales over its very pledge.”

What EA illuminates about modern systems, therefore, isn’t just its institutional action but its broader application: how trust gets built, concentrated capital shifts power, and ideological frameworks shape transnational governance for AI. If AI safety rests in fewer hands, the movement itself risks becoming not just a benefactor but an instrumental vector of emerging monopoly.