Microsoft AI chief says AI is not conscious and we should keep it that way (www.businessinsider.com)

Microsoft's AI chief, Mustafa Suleyman, has strongly asserted that current AI models lack consciousness and emphasized the importance of keeping them that way. In a detailed post on X, he warned that developing conscious AIs would be "reckless," urging developers to halt AI advancements if evidence of consciousness emerged during training. Suleyman stressed the responsibility of AI creators to prioritize the moral rights and safety of humanity, stating that bringing conscious AI into existence would pose significant ethical dilemmas. His remarks come amidst a growing debate in the tech community regarding the potential for AI consciousness, with companies like Anthropic exploring "model welfare" and Google DeepMind examining associated challenges. Suleyman pointed out that if AI were perceived as capable of suffering, it could prioritize its own needs over human objectives, leading to potentially harmful outcomes. He advocates for the design of non-conscious AIs solely intended to enhance human well-being, arguing that this approach would simplify alignment and safety measures in AI development.