The Trump administration has made effective altruism a target in its push to keep American AI development moving quickly. On Sept. 14, a Pentagon technology account posted “Americanism, not effective altruism.” Adviser David Sacks also accused EA-linked groups of promoting fears that could hold the U.S. back against China, Business Insider reported.
The criticism reached other government accounts. The Federal Communications Commission posted a pro-AI frontier-themed graphic, according to Semafor’s Sept. 17 reporting. The messaging portrays caution about advanced AI as an obstacle to technological progress.
Effective altruism, usually shortened to EA, has helped supply AI-safety research with funding and people. But neither the movement nor the wider debate over AI fits neatly into the government’s slogan.
How a philosophy of giving became involved in AI
EA starts with a practical question: how can people use limited resources to help others as much as possible? The movement’s introductory essay describes an approach to comparing problems and interventions, rather than a commitment to one cause. Its examples include malaria prevention, pandemic preparedness and animal welfare alongside AI research.
Some supporters give particular weight to protecting future generations. For them, the possibility of losing control of highly capable AI can justify work on a danger that has not yet materialized. AI alignment research seeks to keep increasingly capable systems acting in accordance with human values, the essay explains. That is a research goal, not evidence that the worst-case outcome is inevitable.
Business Insider describes how EA-linked funders have supported alignment research and organizations that evaluate advanced models. Its analysis places the political dispute around calls to slow the most advanced AI development until safeguards catch up. Officials opposing a slowdown argue that China could gain ground; that is their argument about the consequences, not a demonstrated outcome.
AI safety has a broader constituency
Opposition to unchecked AI development includes people who do not identify as effective altruists. Michael Kleinman, the Future of Life Institute’s head of U.S. policy, told Semafor that his organization is not an EA organization.
The institute helped organize a Pro-Human Summit featuring Sen. Bernie Sanders and former Trump adviser Steve Bannon, Semafor reported. Those participants illustrate why a single philosophical label cannot describe everyone calling for more control over AI development.
