Non-State Actors Access to AI Weapons May Be Unstoppable, UK Parliament Hears

In a hearing before the UK Parliament, James Black, assistant director of defense and security research group RAND Europe, told the House of Lords' AI in Weapon Systems Committee that it would be impossible to contain and curb the proliferation of AI weapons systems among non-state actors such as terrorist groups or mercenaries due to the software-based nature of AI models. Black also said there was no established "war game" theory on how hostile non-state actors might behave using AI-based weapons. The private sector, particularly large multinational corporations, are way ahead of government research in developing this technology. Governments also have no theory on how nefarious groups might behave using the tech. Legislation designed to curb proliferation may not be effective, as there would be a huge incentive to cheat on regulation if these technologies confer profound military advantage. Kenneth Payne, professor of strategy at King's College London, expressed skepticism about the prospects for regulation. Computer scientists and AI experts have recently signed an open letter calling for a pause for at least six months in the training of AI systems more powerful than GPT-4.