AI and Dream Interpretation

Kelly Bulkeley

Can a computer program interpret your dreams? If it could, would that be a good or bad thing? Powerful new systems of artificial intelligence (AI) have raised urgent, destabilizing questions in many areas of life, including dream research. Is an AI tool for interpreting dreams possible? If so, will it help people better understand the meanings of their dreams, or will it become another way in which technology is taking control of our lives? Let’s consider each of these questions in turn.

Is it possible?

The first question is whether a truly automated system of dream interpretation can be built—whether any machine will ever be able to tell you what your dreams mean. I would say no, for the same reason that I’d say no person can tell you, with infallible accuracy, what your dreams mean. Interpreting a dream is ultimately about the dreamer’s own felt sense of inner meaning and significance. A human or a machine can offer the dreamer possibilities, but only the dreamer can know for sure if those possibilities have any actual relevance to personal experience.

In this sense, an AI for dream interpretation could function as a really good symbol dictionary—describing everything the dream might be about, but always leaving it to the dreamer to determine which if any of these potentials have real significance in their lives. No matter how powerful the technology becomes, this basic limit will remain.

Will it help?

Within the context of that limitation—suggesting what a dream might mean, not defining what it does mean—there can be many positive uses for AI-enhanced tools in helping people understand their dreams better. For example, if an AI system were properly trained on the texts of myths, folktales, and sacred stories from around the world, it could presumably provide a dreamer with a Jungian-style amplification of the images and themes appearing in any given dream. Amplification by itself is not interpretation; it goes in the opposite direction, by trying to make connections with symbolic phenomena in the world beyond the dreamer’s personal life. But Jung’s method of amplification is a powerful technique that can stimulate new insights and perspectives, and AI systems may be able to play a helpful role in that process.

For the cultural study of dreaming, which does not rely on personal associations from the dreamers, AI systems may be especially transformative. With this kind of approach, the goal is to identify meaningful connections between a set of dreams and the collective conditions (social, historical, political, economic, religious, artistic, ethical) in which the dreams emerged. It can be very difficult work trying to identify these culturally meaningful connections, and this is where a well-trained AI system could help. Human researchers are still needed to set the analyses in motion, verify their results, and evaluate their significance. But if deployed sensibly, AI systems can provide a tremendous analytic boost to the social scientific study of dreaming.

Can it be abused?

An AI system for interpreting dreams could be abusive if it were treated as a higher psychological authority; as an oracle of truth rather than a helpful tool. That’s a genuine danger, and one to guard against by always emphasizing the irreducibly human element in interpreting the personal meanings of dreams.

Perhaps future advances will address this concern, but I wonder how well any technology can be designed to account for the radical openness of dreaming. Dreaming is defined by its chaotic, non-linear qualities—by its novelty, playfulness, spontaneity, randomness, and bizarreness. Will an AI system trained on ordinary linguistic texts be able to avoid homogenizing and domesticating the unpredictable wildness of dreaming? Will it ever be able to recognize the unique serendipities that emerge in dreams, the startling synchronicities, the creative flashes, the shocking transgressions, the uncanny anomalies? I have my doubts.

That being said, the impact of new generations of AI is not (yet) as frighteningly disruptive for dream interpretation as they have been for other areas of human activity. AI systems seem likely to become dream dictionaries on steroids—highly detailed and richly suggestive, and yet always requiring the dreamer’s own reflections to determine the actual meanings.