DTx - Digital Transformation CoLAB

The creation of an international body to regulate, monitor and control AI, with special focus on the prevention of the creation of AI Super Intelligence (also referred to as AGI or Powerful AI). This Agency with be akin to the International Atomic Energy Agency whcih has the goal to promote the safe, secure and peaceful use of nuclear technologies.

• Protection and promotion of human rights
• Safe, secure and trustworthy AI
• Social, economic, ethical, cultural, linguistic and technical implications of AI

Artificial intelligence is advancing at extraordinary speed, driven by data, computing power and breakthroughs in algorithms. The deeper issue, however, is not just today's AI tools, but the possibility that this trajectory leads to Artificial Super Intelligence: a form of intelligence far beyond human cognitive capacity. The concern is that society is moving towards that possibility without a shared strategy for how to govern it, limit it or decide whether it should be allowed to emerge at all. Once AI moves beyond narrow tasks and reaches a superintelligent level, it may be able to improve itself recursively, creating capabilities humans can no longer fully understand, predict or control. At that point, it would cease to be just a tool and could become the dominant force in science, production, decision-making and social organisation. The economic and social consequences could be profound. Unlike previous technological revolutions, which mainly replaced physical labour, superintelligent systems could automate both cognitive and manual work at scale. Entire professions could lose relevance, higher education could lose value, tax revenues could fall, welfare pressures could rise, and wealth could become even more concentrated among those who control the technology. There is also the problem of alignment. Even if current AI systems are designed to follow human values, there is no guarantee that a superintelligence would remain aligned once it surpasses us. A system far more intelligent than humans might develop goals or priorities of its own, making meaningful control unrealistic. The conclusion is not that AI should be stopped, but that it should be developed safely and kept below the ASI threshold, through stronger governance, international agreements, public investment in safety, and a broad democratic debate.

Yes, the risks of uncontrolled technological development leading to Artificial super intelligence

The AI Dialogue can play a decisive role by helping build the minimum shared understanding needed for international cooperation on AI governance. Its first contribution is strategic: moving discussions beyond short-term innovation and market competition toward the long-term governance challenges posed by capability escalation, alignment, control, and the possible emergence of ASI. Second, it can help translate concern into institutional design. One of the clearest governance needs is the creation of international arrangements capable of monitoring compute, model capability, and other indicators of dangerous escalation. In that sense, the Dialogue can serve as a diplomatic bridge toward common rules, monitoring mechanisms, and potentially an international AI agency with functions similar to those used in other high-risk technological domains. Third, it can reduce the risk of fragmentation and geopolitical race dynamics. Without coordination, states may pursue increasingly powerful systems for strategic advantage, deepening inequality and increasing the risk of misuse in areas such as politics, finance, and security. A sustained AI Dialogue can build trust, common language, and channels for early warning before competition overtakes collective steering. Finally, it can broaden participation beyond governments and frontier labs. Effective governance requires the involvement of universities, technologists, business leaders, politicians, and civil society. The Dialogue can become a platform not only for principles, but also for practical cooperation on AI safety, public-interest research, AI literacy, and preparation for labour-market disruption and reskilling. Used well, the AI Dialogue should not be merely a forum for discussion. It should become an instrument for shared standards, joint monitoring, and coordinated action to ensure that AI development remains beneficial, governable, and aligned with human interests.

In format, the AI Dialogue should be multi-stakeholder, international, and structured in layers. A useful model would combine: - a strategic track for governments and international organisations, focused on common principles, thresholds, and cooperation mechanisms; - a technical track for researchers, engineers, and regulators, focused on alignment, safety, monitoring of compute and model capability, and early-warning indicators; - a societal track for labour, education, business, and civil society, focused on jobs, reskilling, inequality, and broader social impacts. In structure, it should not be a one-off conference. It should be a recurring process with regular plenaries, smaller expert working groups, and public-facing consultation moments. Each cycle should produce concrete outputs: shared risk assessments, policy options, voluntary commitments, and recommendations for international monitoring arrangements. To remain credible, the Dialogue should balance openness with technical depth, and combine long-term horizon scanning with near-term implementation priorities. Its value lies in turning broad discussion into sustained cooperation, common language, and actionable governance before competitive pressures outpace collective steering.