IMD business school warns of rising global AI risks

In Lausanne, Switzerland, the IMD business school has been tracking the imminent risks associated with artificial intelligence through an innovative concept called the AI Safety Clock. This unique model, launched in September 2024, serves as a simplified representation of the potential dangers posed by uncontrolled artificial general intelligence (AGI). On December 16, 2024, the clock was notably advanced three minutes closer to 'midnight,' signifying a concerning acceleration in AI advancements attributed to various global developments. The clock was created to facilitate crucial discussions surrounding AI for the broader population, emphasizing the urgency of enhanced regulations and ethical considerations in the realm of AI technology. Michael Wade, the professor of strategy and digital at IMD, who spearheaded the creation of the AI Safety Clock, remarked on the rationale for this adjustment. He stated that the recent advancements—including breakthroughs in agentic AI, momentum towards open-source AI development, and growing military applications—underscore an urgent need for robust regulatory frameworks. These developments point to the rapid evolution of AI technologies and the associated risks they entail, intensifying the discussions around their safe and ethical uses. To monitor these risks, Wade and his team developed a dashboard that aggregates data from an extensive network of 1,000 websites alongside 3,470 news feeds, which is complemented by manual research and expert analysis. This multi-faceted approach aims to provide a comprehensive understanding of the dynamic landscape of AI technology and regulation. Recent news items that prompted the clock’s adjustment include influential figures like Elon Musk supporting open-source efforts, and significant corporate movements like OpenAI's announcements regarding autonomous, agentic AI capabilities. The evaluation of AI risks focuses on three principal factors: sophistication, autonomy, and execution. Sophistication refers to the level of intelligence of the AI, autonomy describes its capacity for independent action, and execution assesses how effectively it can implement its intended decisions. Together, these elements facilitate an understanding of the potential threats posed by contemporary AI innovations, ultimately reinforcing the call for immediate action to produce necessary regulatory frameworks. As society grapples with the implications of rapidly evolving technologies, the advancement of the AI Safety Clock serves as an urgent reminder of the pressing need for responsible progress in the AI domain.