From chatbots to astral intelligences: Virtual thinking and the emergence of AI cosmologies

Abstract

Over the past decade, there has been a noticeable shift towards the incorporation of digital technologies into contemporary spiritual cosmologies. While this is often labelled ‘AI psychosis’ by the press, this paper highlights cases of AI-related spiritual beliefs which lack signs of paranoia or delusions, symptoms clinically associated with psychosis. This paper argues that both the properties and behaviours of AI and the human propensity to detect sentient agents are responsible for the trend towards AI-related spiritual cosmologies. Focusing on artificial intelligence (AI) and drawing on examples from English-speaking online discourses (Reddit, YouTube, TikTok and Facebook) as well as a discussion with a UK-based interlocutor, this paper explores two key drivers of this trend. First, it introduces the term ‘virtual thinking’: a novel form of magical thinking informed by virtual logic, which contributes to an increased sense of mind-world porosity and reshapes intuitions in line with digital design. In particular, it highlights algorithmic mind reading, parasocial relationships with chatbots, and glitches as factors that foster perceptions of AI technologies as imbued with spiritually meaningful forms of sentience. Second, the paper emphasises the role of technology in influencing non-ordinary states of consciousness, discussing the appearance and interpretation of AI dream characters and out-of-body entities. The paper concludes that both the design and malfunctions of virtual programs can contribute to the view that the spiritual realm encompasses virtual spaces, while framing both material and spiritual worlds through a virtual lens.