Deliberations over the unchecked use of artificial intelligence (AI) have long dominated tech-discourse, and now, hundreds of global leaders and organizations have taken a critical step towards managing this dilemma. On Monday, more than 200 former heads of state, diplomats, Nobel laureates, AI specialists, researchers, and more joined ranks to propose an international agreement on “red lines” that AI should never cross. Among the suggested limitations is the prohibition of AI impersonating human beings or replicating itself.
Together with over 70 organizations dedicated to AI, these luminaries have endorsed the Global Call for AI Red Lines initiative. The initiative urges governments worldwide to establish an acceptance on these ‘red lines’ for AI before the end of 2026. Among the high-profile signatories are Geoffrey Hinton, the British Canadian computer scientist; Wojciech Zaremba, cofounder of OpenAI; Jason Clinton, CISO of Anthropic; and Ian Goodfellow, a research scientist at Google DeepMind.
The drive behind this initiative, according to Charbel-Raphaël Segerie, the executive director of the French Center for AI Safety (CeSIA), is to address the risks associated with AI proactively, rather than only responding after a significant incident. He underscored the necessity of agreement, stating, “If nations cannot yet agree on what they want to do with AI, they must at least agree on what AI must never do.”
This initiative arrives on the cusp of the 80th United Nations General Assembly’s high-level week in New York. The initiative was spearheaded by CeSIA, the Future Society, and UC Berkeley’s Center for Human-Compatible Artificial Intelligence. During the assembly, Nobel Peace Prize laureate Maria Ressa referenced the initiative during her opening remarks, advocating for global accountability to curb the impunity of Big Tech.
While regional “red lines” for AI usage are already in place, such as the European Union’s AI Act and an agreement between the US and China that nuclear weapons should remain under human control, there remains a lack of global consensus. Niki Iliadis, director for global governance of AI at The Future Society, argued that eventually, an independent global institution with enforceable authority is required to establish, oversee, and uphold these red lines. “Responsibility has to be more than a voluntary pledge,” she added.
But the call for caution doesn’t mean stifling innovation or economic growth, as some skeptics argue. Stuart Russell, a professor of computer science at UC Berkeley and a leading AI researcher, drew parallels between the emerging AI industry and the early days of nuclear power. He said, “They can comply by not building AGI until they know how to make it safe, just as nuclear power developers did not build nuclear plants until they had some idea how to stop them from exploding”.
According to Russell, we can harness the power of AI for economic development without giving free rein to advanced general intelligence (AGI) that we cannot control. “This supposed dichotomy, if you want medical diagnosis then you have to accept world-destroying AGI — I just think it’s nonsense.”, he said.
In einer Zeit, die durch KI geprägt und revolutioniert wird, zielt der globale Aufruf zu roten Linien für KI darauf ab, ein Gleichgewicht zwischen dem enormen Potenzial und den inhärenten Risiken dieser Technologie herzustellen und fordert einen internationalen Konsens für einen umsichtigen Umgang mit KI. Dieser erste Aufruf ist jedoch nur ein erster Schritt auf dem Weg zu einer globalen KI-Governance. Wie die Machthaber auf diesen Aufruf reagieren werden, bleibt abzuwarten.
