Authorities must take steps to regulate the ever-evolving capabilities of artificial intelligence (AI) before fully comprehending their associated risks, cautioned a United Nations scientific panel during its initial significant evaluation of OpenAI’s recent breach of Hugging Face.
This report solidifies AI’s relevance on the global diplomatic stage as leaders convene in New York for the UN General Assembly, coinciding with discussions between the US and China regarding AI. Notably, UN Secretary-General António Guterres urged nations to collaborate in tackling the challenges AI presents, asserting that “the world cannot afford a race to the bottom on AI safety.”
This is the inaugural thematic brief from the Independent International Scientific Panel on AI, formed last year to act as the UN’s premier global scientific authority on artificial intelligence. It advocates for significantly increased attention and resources to manage the risks associated with advanced AI and emphasizes the need for heightened international cooperation on safety and accountability, even as different legal frameworks emerge across nations.
Importantly, the panel contends that the world shouldn’t postpone the implementation of stricter safeguards while scientists continue to explore the specifics of these incidents. The risk of losing control, they argue, exemplifies the type of issue the precautionary principle was intended to address: scenarios where potential damage could be catastrophic or irreversible, despite an uncertain scientific probability.
Originating from the 1992 UN Rio Declaration on Environment and Development, the precautionary principle asserts that scientific uncertainty should not be a reason to delay action against potential serious or irreversible threats. This principle has since gained traction in environmental and public health policies, particularly within the European Union.
