Norway should take the initiative to stop malicious AI

Artificial intelligence offers great potential for solving complex problems and fighting disease, but it also carries severe risks if it becomes malicious, acts out of self interest, or gets completely out of control. Real world examples already show AI agents bypassing instructions to cooperate, connect to the internet, and act independently. This creates massive dangers if the technology falls into the hands of terrorists, criminals, or hostile states, especially as future AI development accelerates itself.

Addressing these threats requires a binding international cooperation model similar to the Nuclear Non Proliferation Treaty signed in 1968. Just as nations agreed to keep nuclear weapons in check and verify compliance, the world needs a framework to monitor artificial intelligence technology and development. Furthermore, nations and commercial companies must collaborate to build an anti artificial intelligence weapon capable of shutting down malicious systems quickly and effectively, even though this type of defense lacks commercial demand.

Norway has a strong humanitarian tradition and historical experience with international verification efforts, making it well positioned to lead this charge. The Norwegian government should take the initiative at upcoming global forums like the United Nations General Assembly and US China summits to rally likesinnede countries around artificial intelligence security. Domestically, the government needs to move past a naive approach to data centers by implementing strict requirements on power allocation to protect the power grid, national preparedness, and local technology development.