According to an advisory group with the United Nations, the growth of artificial intelligence (AI) is outpacing our ability to control it, and the group wants the UN to start working on how to regulate the technology.
In a 100-page report released September 19, the group said AI is transforming the world and has the potential to greatly advance scientific research and economic growth. The technology also holds promise for public health, farming, and the management of electricity grids, it said.
But without a regulatory structure, the advisory board said that the benefits of AI would only be available to the richest countries and businesses. At the same time, if it grows more powerful, AI could “upend the work of work.” The group is also worried that the technology could become sophisticated enough to create and design its own weapons.
The report articulated foundational principles the group believes should be adhered to when new governing bodies are formed to oversee AI. Critical areas of focus included taking account of international law and human rights law. How this will be accomplished across nations is difficult to predict, as countries around the world have very different ideas about what constitutes “human rights,” and some seem not to believe the concept at all.
Among the recommendations in the AI Advisory Body report are the establishment of a global scientific panel to assess the risks and benefits of AI. The group also wants a “dialogue” at the UN on how the world body can make sure that any AI regulation adheres to human rights principles.
The report also calls for a global fund to study AI with a focus on how to ensure that the technology is available to both rich and poor nations.
Right now, according to the report, only seven of the 193 countries in the UN have signed onto any of the group’s AI initiatives. It said the remaining 118 are part of the “Global South” and they are “missing entirely” from the discussions about AI.
One of the first such initiatives is a European Union framework to regulate AI, which went into effect August 1.
