South Korea is gearing up to host the second global AI safety summit on May 21-22, following Britain's inaugural event at Bletchley Park six months prior. Government officials announced that the summit would build upon the "Bletchley Declaration," an agreement among nations, including the U.S. and China, to collaborate on AI safety.
The decision to convene these summits stems from concerns raised when Microsoft-backed OpenAI released ChatGPT in late 2022. While many marveled at the technology's capabilities, others, like former Google researcher Geoffrey Hinton and Elon Musk, voiced apprehensions, with Hinton suggesting AI posed a more immediate threat than climate change.
In response, Britain hosted the first AI Safety Summit in November, drawing leaders, tech luminaries, and scholars to devise regulatory frameworks. Following the event, over 25 government representatives signed a joint declaration pledging international cooperation and unified oversight approaches.
Given the rapid evolution of AI, future summits are scheduled every six months. South Korea will primarily host a virtual event in May, with France slated to organize the subsequent in-person summit later in 2024.
Lee Jong-Ho, representing South Korea's Ministry of Science and ICT, underscored the significance of the transition from Britain to South Korea, emphasizing the need for global norms and governance to guide the responsible deployment of AI technologies for the betterment of humanity.