The Alliance for Peacebuilding’s Call to Action: Regulating AI is a Global Effort
FOR IMMEDIATE RELEASE
Last updated September 29, 2026
CONTACT
Nick Zuroski | nick@allianceforpeacebuilding.org
Washington DC, USA — “Human beings built these systems, and human beings can decide to build them in ways that don't ‘kill us all.’ AfP is a part of the effort to design systems that don't divide and polarize us toward conflict.”
-Ravi Iyer, Managing Director of USC Marshall School’s Neely Center for Ethical Leadership and Decision Making and the Psychology of Technology Institute
The Alliance for Peacebuilding (AfP) calls for strong, coordinated action across governments, the United Nations, and technology companies in partnership with peacebuilding organizations so AI promotes peace, rather than chaos. Last week, Anthropic researcher Jacob Coxon (also formerly with OpenAI) resigned, claiming “neither company is acting responsibly” and “gambling with our lives.” In the words of UN Secretary-General António Guterres, “The world cannot afford a race to the bottom on AI safety.” But it is not too late to act.
The United Nations must use its convening power to outline principles and best practices, including by building off of the work of its AI Advisory Body. The UN’s AI for Good framework and the recent Pact for the Future and Global Digital Compact are starting points, but must go farther. The UN needs a dedicated, technically capable body with real standing, similar to the International Atomic Energy Agency (IAEA) for nuclear weapons, as the current advisory bodies have no teeth. This kind of body is critical, given U.S.-China strategic competition that creates powerful incentives for both states to race, rather than regulate, as each state fears falling behind in technological advancement. Member States must agree to something like a Permanent AI Safety Council or, at a minimum, a standing intergovernmental panel (like the Intergovernmental Panel on Climate Change) that tracks AI risks, including autonomous weapons, disinformation infrastructure, and AI-enabled surveillance of minorities and dissidents. In the short term, the UN can also use existing mechanisms—the Human Rights Council, the Office of the United Nations High Commissioner for Human Rights, special rapporteurs—to document and name AI-enabled human rights violations, which creates accountability pressure, even without enforcement.
Governments worldwide must step up and use their direct leverage. Domestic regulation is lagging badly. The European Union’s AI Act established risk-tiered regulation, with high-risk applications (biometric surveillance, critical infrastructure, social scoring) facing strict requirements; the U.S. has no equivalent.
The U.S. Congress has been slow to respond to increasing AI concerns and threats. While we welcome recently introduced and proposed bipartisan legislation, Congress must urgently hold oversight hearings and adopt comprehensive legislation, which has stalled across the board.
In the Senate, proposed legislation includes:
-Senate Majority Leader John Thune (R-SD) and Senators Amy Klobuchar (D-MN) and Ted Cruz (R-TX) are drafting legislation—which has not yet been unveiled—that would require AI developers to work with the government to verify and test models to ensure safety before models are released. The bill would also require developers to report any imminent risks associated with their products.
-Senators Jim Justice (R-WV) and Mark Kelly (D-AZ) introduced the Senior Chatbot Protection Act: a bipartisan bill to protect older adults from AI scams, deceptive designs, and unsafe chatbot interactions.
-Senator Ron Wyden (D-OR) introduced the Algorithmic Accountability Act, which would require bias testing and impact assessments for high-stakes AI.
-Senator Josh Hawley (R-MO) introduced the AI Risk Evaluation Act, which would mandate federal review of high-impact AI systems.
-Senator Chuck Grassley (R-IA) introduced the AI Whistleblower Protection Act, which would shield workers who report AI safety violations.
-Senator Ed Markey (D-MA) introduced the States' Right to Regulate AI Act, explicitly pushing back against a December 2025 Executive Order that limited state authority over AI. Senator Brian Schatz (D-HI) introduced The GUARDRAILS ACT that would cancel that same executive order and block its funding. Similar legislation has been proposed in the House of Representatives by Representative Don Beyer (D-VA)
-Senators Marsha Blackburn (R-TN), Amy Klobuchar (D-MN), and Thomas Tillis (R-NC) introduced the NO FAKES Act, which would create a federal intellectual property right to protect individuals from unauthorized digital replicas of their voice and visual likeness.
-Senator Bernie Sanders (I-VT) introduced the AI Data Center Moratorium Act, which would pause new data centers pending an environmental study, and announced work on the Ban Artificial Superintelligence Act, forthcoming legislation that would prohibit AI that surpasses human intelligence with prison sentences of up to 20 years.
-Senator Dick Durbin (D-IL) introduced the Data Center Water and Energy Transparency Act, which would require disclosure of AI data centers' water and energy use.
-Anticipated legislation by Senator Mark Kelly (D-AZ) would create a fund AI companies would pay into to support workers whose jobs could be affected by the technology, helping them with job training, enhanced unemployment benefits, and starting their own businesses.
In the House of Representatives, recent activity includes:
-Representative Ro Khanna (D-CA), who represents part of Silicon Valley, has pushed for extensive AI regulation and sent letters to three top Chinese AI labs asking them to work with leading U.S. tech companies to “pace the frontier.”
-Representatives Nick LaLota (R-NY) and Josh Gottheimer (D-NJ) announced the China FIREWALL Act, which would ban all Chinese-developed open-weight AI models from government-issued devices and prohibit federal agencies from procuring software that relies on those models.
-Representatives Josh Gottheimer (D-NJ) and Mike Lawler (R-NY) announced the Stop Rogue AI Act, which would direct the National Institute of Standards and Technology (NIST) to develop national standards, guidelines, and best practices for discovering, verifying, monitoring, and controlling AI agents.
-Representatives Sam Liccardo (D-CA) and Kevin Kiley (I-CA) announced the AI Safety & U.S.-China Standards Bill, which would direct the State and Commerce departments to pursue binding U.S.-China safety standards, independent auditing, intellectual property protections, and a direct government-to-government "red phone" hotline for urgent AI risks.
The U.S. Administration should issue an Executive Order that sets safety standards for federal procurement and use. We welcome recent comments from Treasury Secretary Scott Bessent that the U.S. is proposing a notification system between Washington and Beijing that would allow the U.S. and Chinese Governments to contact one another in the event of an AI-related incident with national security implications.
The U.S. military must establish internal governance guidelines. According to recent reports, a faulty AI-generated intelligence report earlier in 2026 nearly triggered an armed U.S. military operation against a Chinese ship in the Middle East. We welcome Senator Mark Kelly’s proposed legislation—including the Ultimate Human Responsibility in Defense Systems Act and the Warfighter AI Readiness and Preparedness (WARP) Acton—that would seek to safely integrate AI into the U.S. military and require the Department of Defense to assess the impact of AI in the military sector and keep humans in command of decisions involving the use of force.
Multilateral coordination is desperately needed. The Bletchley Declaration and subsequent AI Safety Summits established a forum, but they are voluntary and industry-heavy. Governments must call for a binding treaty framework among democracies, starting with the G7, that establishes shared standards for high-risk AI, including disinformation, autonomous weapons, and AI-enabled election interference.
Civil Society must use its independent agency to provide more pressure and accountability to call for regulating AI. International norm-building has historically been driven significantly by civil society. The Women, Peace, and Security Agenda, the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on their Destruction, the Arms Trade Treaty, and the campaign against cluster munitions were all civil society-led. Civil society, especially peacebuilding and prevention organizations, must robustly engage with technology companies and policymakers to embed peacebuilding best practices across the entire AI design cycle. This partnership model ensures the inclusion of communities most impacted by violence, and that technologists can develop frameworks and standards that prevent, reduce, and mitigate AI-enabled conflict.
The peacebuilding field has an opening to engage companies directly on conflict-specific harms by bringing documented evidence of how AI tools have contributed to violence or atrocities, proposing specific design changes, and making the case for conflict-sensitive AI. And civil society brings a wealth of expertise, informed analysis, and recommendations it can draw on for this work, including from the Digital Peacebuilding community, as well as the Council on Technology and Social Cohesion (which AfP leads two hubs of) and its mission to design technology that unites people instead of dividing them. We must not give in to panic—only calling out problems—and instead be part of a whole-of-society effort that minimizes the technology’s harmful impacts and maximizes its contributions to peacebuilding innovation.
We welcome calls from technology companies to pause AI development and avoid releasing new models amidst concerning “autonomy and deception” by AI agents, as these companies have the most capacity to address AI harms, but also have the strongest financial incentives not to. Beyond pausing development, technology companies must urgently integrate conflict prevention into their development of AI and hire conflict experts that can identify risks and develop responses as the technology grows. Companies will also have to cooperate with each other instead of competing. It is not hyperbole to compare this technology to the arms control race, and similar safeguards must be put in place across governments and companies building AI.
The Alliance for Peacebuilding (AfP), named the “number one influencer and change agent” among peacebuilding institutions worldwide, is an award-winning nonprofit and nonpartisan network of almost 300 organizations working in 181 countries to prevent and reduce violent conflict and build sustainable peace. AfP cultivates a network to strengthen and advance the peacebuilding field, enabling peacebuilding organizations to achieve greater impact—tackling issues too large for any one organization to address alone.