Abstract

The article explores the phenomenon of consciousness through the lens of advances in artificial intelligence (AI) and of contemporary neurobiological theories, each offering a distinct account of the architecture of conscious processing. This theoretical landscape allows us to identify several specific properties – such as a global workspace, recurrent processing, metacognitive monitoring, predictive processing, and high-level information integration – as functional signatures of cognitive processes inherent to biological consciousness. While certain AI systems exhibit some of these properties, they currently manifest in a fragmented and poorly integrated manner. The transition toward hybrid, particularly neurosymbolic, architectures, coupled with the expanding use of neuroevolutionary and embodied approaches in robotics, is laying the groundwork for integrated systems that more closely approximate conscious cognitive functions. However, the necessary conditions for a genuine dialogue between humans and a potentially conscious technosubject – including elements of intersubjectivity, empathy, mutual ethical responsibility, and lived bodily and social experience, or at least functional analogues thereof – suggest that the capacity for such interaction transcends the mere simulation of functional properties. The potential emergence of artificial general intelligence (AGI) in the near future, an entity capable not only of performing all human cognitive functions but also of demonstrating autonomous behavior and engaging in genuine dialogue, necessitates a proactive discussion of the technosubject’s normative status (the bounds of agency and responsibility, rights, duties, and safeguards for humans), along with the development of appropriate ethical principles and regulatory mechanisms.