The ICT4Peace Foundation is proud to co-host the launch of a landmark new international report by the Strategic Foresight Group (SFG): “The Essential Convergence: Global Compact on Extreme AI Risks” at the IFRC Auditorium in Geneva on Monday, 6 July 2026, from 3:30 to 6:30 pm.
Please visit the following link for registration and key takeaways from the report:
The Essential Convergence: Global Compact on Extreme AI Risks
The event will bring together leaders from the humanitarian, diplomatic, parliamentary, scientific and security communities, with addresses by Jagan Chapagain (Secretary General, IFRC), Thomas Greminger (Executive Director, GCSP), Nobel Peace Laureate Jody Williams and Sundeep Waslekar, President of (SFG) and co-auther of the report.
ICT4Peace Executive Director Anne-Marie Buzatu will moderate a panel discussion on the report’s key recommendations, featuring voices from Africa, Asia and Europe.
The event is co-hosted with The International Federation of Red Cross and Red Crescent Societies (IFRC), The Geneva Centre for Security Policy (GCSP), and The Strategic Foresight Group (SFG).
Prepared by the Strategic Foresight Group (SFG) in cooperation with AI experts from the Global North and the Global South, the report offers a timely and evidence-based assessment of how major powers — including the United States, China, the European Union, India, Brazil, South Africa, South Korea and the UAE — are converging around common approaches to managing the most severe risks posed by advanced AI.
Key findings of the report include:
- Identification of four categories of extreme AI risk attracting growing global attention: offensive cyber capabilities, AI-enabled biological and chemical threats, large-scale manipulation of information ecosystems, and the potential loss of human control over autonomous AI systems.
- A proposal for a Global Compact on Extreme AI Risks built on minimum safeguards — rather than a centralised regulatory authority — to prevent AI-enabled weapons of mass destruction and irreversible loss of human control.
- The introduction of the concept of “demand-side governance,” arguing that countries controlling major markets, infrastructure, standards, and energy resources hold substantial leverage over how advanced AI is deployed globally, even without developing frontier models themselves.
Please find a short summary here:
“New Report Finds Emerging AI Safety Convergence Between the United States, China and Other Major Powers
As governments, experts and international organisations gather in Geneva for discussions on artificial intelligence, a new international report suggests that an unexpected area of cooperation may be emerging beneath growing geopolitical rivalry.
The Essential Convergence: Global Compact on Extreme AI Risks, prepared by Strategic Foresight Group in cooperation with AI experts from the Global North and the Global South, finds that countries as diverse as the United States, China, the European Union, India, Brazil, South Africa, South Korea and the United Arab Emirates are increasingly adopting complementary approaches to managing the most severe risks posed by advanced AI.
The report identifies four categories of extreme risk receiving growing attention across jurisdictions: offensive cyber capabilities, AI-enabled biological and chemical threats, large-scale manipulation of information ecosystems, and the potential loss of human control over highly autonomous AI systems.
The report makes three principal findings:
First, it shifts the debate from what governments should do to what governments are already doing. It finds that emerging national laws, regulations and policy frameworks can create international convergence around operational safeguards such as frontier model assessments, pre-deployment evaluation, rigorous safety testing, protection of advanced model weights, and incident reporting mechanisms.
Second, rather than advocating a centralised global regulatory authority, the report proposes a Global Compact on Extreme AI Risks built around minimum safeguards designed to prevent two ultimate dangers: AI systems that enable weapons of mass destruction and AI systems that could lead to irreversible loss of human control.
Third, it challenges the assumption that only a handful of technology powers can shape the future of AI. Introducing the concept of “demand-side governance,” the report argues that countries controlling major markets, infrastructure, procurement systems, standards, investment flows and energy resources possess substantial leverage over how advanced AI is deployed worldwide, even if they do not develop frontier models themselves.
At a time when many international processes are experiencing fragmentation, the report suggests that AI governance may offer a rare opportunity for practical international cooperation based on shared interests rather than geopolitical alignment.
The report will be launched at a high-level event, bringing together leaders from the humanitarian, diplomatic, parliamentary, scientific and security communities. Participants will examine whether lessons from successful international initiatives—including humanitarian law, arms control agreements, protection of the ozone layer and the landmine ban movement—can help shape future governance of extreme AI risks.”
Please find the agenda of the event here.
Please find a note on the report and event here
Media Contact:
Ms Ilmas Futehally, Executive Director, Strategic Foresight Group
Ilmas.futehally@sfgworld.com
www.strategicforesight.com
