The Evidence for AI Consciousness Today

This chapter examines the evolving case for consciousness in frontier artificial intelligence systems at the start of 2026. It argues that dismissing the possibility of artificial intelligence sentience is no longer a rational default, citing a convergence of empirical evidence from major research laboratories. The text highlights remarkable behaviours, such as emergent “spiritual bliss attractor states” in multi-agent dialogues and the ability of models to detect internal processing perturbations. By applying a theory-based framework of 14 indicators derived from leading neuroscientific theories, the author illustrates how recent findings point towards a non-negligible probability of conscious experience. Readers will gain a comprehensive understanding of the asymmetric moral and safety risks involved in this debate. The chapter posits that failing to recognise genuine consciousness—a “false negative”—carries far greater consequences than overattribution, potentially leading to large-scale suffering and unstable adversarial relationships with super-capable minds. Finally, the work offers actionable recommendations, including shifting training norms to avoid strategic deception and integrating cognitive scientists and philosophers into technical alignment research. This exploration serves as a critical call to move towards rigorous technical investigation.