UN leadership?
In 2023, the UN launched a 39-member advisory body to examine AI’s international governance. Now, following the release of its report, the advisory body is calling for the establishment of a panel to provide impartial scientific knowledge about AI. This panel will also address information nonuniformity between AI labs and the world.
“In principle, a coordinated worldwide centralised approach to AI regulation would appear to be a good idea – enforcing a consistent approach, which is good for organisations, and increasing competition by keeping a relatively level playing field,” said Peter Galdies, Director, DataIQ. “There are, however, some serious issues to consider. Firstly, the UN itself does not consist of all nation states – and even some of those that are represented do not appear to stick to the principals it encompasses.
“Secondly, it would appear to be appropriate to question the ability of the UN to enforce its own resolutions – how would this play out with bad actor states who do not want to stick to the rules anyway?
“Thirdly, the pace of change in AI is so fast how could a body like the UN react fast enough to evolve with the technology – would this over brake developments that could ultimately benefit everyone?”
The issue with AI regulation on a global scale has been that different geographies and nations have been taking different approaches. The EU has been ahead of the curve by introducing the comprehensive EU AI Act, whereas the US has taken an opt-in approach of voluntary compliance – aside from California’s new laws and Biden’s executive order (which could be repealed after the elections). In Asia, China has taken an approach to maintain social stability and state control with any floated notions of AI laws perceived as taking too long to implement (if implemented at all) and probably ineffective.
One of the main concerns in the AI race is that the development of AI seems to be under the remit of just a few multinational companies, and there is a real fear that the technologies could be imposed on people without them having a say in how it is used. If the UN can take the reins and implement global AI regulations, it could help create an equal playing field for all nations – but it could also risk stifling innovation from those at the cutting edge of AI development with the resources to spare.
In its statement, the UN recommended a new policy dialogue on AI governance where it would create an AI standards exchange and boost governance capacity through a development network. With this in place, it would mean that the UN’s 193 members can discuss risks and agree upon actions, and not be led by markets, individual nations, or single multinational businesses. This would be supported by an AI office set up by the UN.
Additionally, a global AI fund could be launched to address concerns with collaboration and capacity, while bringing a global AI framework to the forefront. It would be hoped that this sort of approach would improve transparency and accountability – two of the recurring concerns that seem to follow AI development.
DataIQ members can measure their compliance and regulation capabilities through this unique assessment.
