Yann LeCun, Meta’s chief AI scientist, discusses the topic of AI regulation. He emphasises the importance of open-source AI as a global asset.
He advocates that countries should avoid hindering open-source platforms. AI has become a prominent topic in diplomatic discussions in recent years, sparking debates among researchers, technology leaders and policymakers on the governance of open-source models, which can be freely modified and used.
At the AI Action Summit in Paris, Yann LeCun presented his vision for training open-source platforms using decentralised data centres around the world. These centres would have access to their own data sources, contributing to a shared model that would represent a comprehensive repository of human knowledge. LeCun suggests that, in order to realise this vision, countries must carefully consider regulations and legislation that favour open-source initiatives.
Similarly, OpenAI CEO Sam Altman stresses the necessity of international regulation for both open-source and closed-loop AI systems. He warns of the potential global harm that advanced AI systems could cause, and calls for oversight by an international agency to ensure the safety testing and regulation of the most powerful systems.
