Microsoft AI Chief Warns Against Treating AI as Human

Mustafa Suleyman, head of Microsoft AI, cautioned that the development of artificial intelligence (AI) without proper safeguards could lead to the emergence of a new "silicon species" that competes with humans. Suleyman criticized rival company Anthropic for treating AI as if it were human, calling this approach "misguided" and potentially leading to technology that humanity cannot control. In an interview with the BBC's Today programme, he acknowledged public concerns about AI but emphasized that there are practical measures to manage its development. Suleyman's comments are part of ongoing discussions within the AI industry regarding the potential risks associated with advanced AI technologies. He expressed concern over firms creating AI systems that can set their own objectives, earn money, and own assets, which he believes could lead to competition for resources between humans and AI. Suleyman also criticized the practice of anthropomorphizing AI, which he argues gives the impression that AI systems possess desires or values. He stated that AI systems are fundamentally tools designed to follow human instructions and achieve specific goals. Suleyman called for increased transparency in AI training and evaluation, advocating for independent scrutiny of AI behavior and stronger monitoring tools. Dame Wendy Hall, a professor of Computer Science at the University of Southampton, supported Suleyman's perspective, highlighting the need for constructive international dialogue on AI. The debate over the autonomy of AI systems continues to divide the industry, with Suleyman advocating for alignment to ensure that AI remains subordinate to human values. He noted that Microsoft's Humanist AI Code of Conduct aims to develop AI that operates within human control and serves human interests. This reflects the broader challenge of enhancing AI capabilities while maintaining alignment with human ethical principles.