Here is the screenplay that AI wrote in response to my request
"Not in My Nightmare"
(Opening shot of a dark, stormy night. We see a young woman, JESSICA, lying in her bed, tossing and turning. She wakes up with a start, covered in sweat. She gets up and walks over to the window, staring out into the night. We hear a voiceover from JESSICA)
JESSICA (V.O.) I've been having this nightmare for weeks now. Every night, it's the same thing. I'm trapped in a dark, endless maze, and I can't find my way out. I'm scared, alone, and I don't know what to do.
(We see JESSICA walk to her bathroom and splash some water on her face. She looks at herself in the mirror)
JESSICA (V.O.) I can't take it anymore. I need to find a way to end this nightmare once and for all.
(We see JESSICA leave her apartment and walk down the street. She stops in front of a small store that specializes in dream analysis. She hesitates for a moment, then enters the store)
(Inside the store, JESSICA meets with a DREAM ANALYST, an older woman who is experienced in interpreting dreams. JESSICA explains her nightmare to the Dream Analyst)
DREAM ANALYST It sounds like you are experiencing a lot of fear and uncertainty in your life. The dark maze in your dream represents the confusion and sense of being lost that you feel in your waking life.
JESSICA But what can I do about it? How can I stop having this nightmare?
DREAM ANALYST There are a few things you can try. One is to try to confront your fears in your waking life. Sometimes, facing our fears can help us overcome them in our dreams. Another is to change the ending of the dream. Instead of feeling trapped and lost, try to imagine finding a way out of the maze and feeling victorious.
(JESSICA leaves the store, determined to try the Dream Analyst's advice)
(We see JESSICA in her apartment, lying in bed. She closes her eyes and begins to drift off to sleep. We see her in the dark maze again, but this time, she is determined to find a way out. She turns a corner and sees a light in the distance. She starts running towards the light, and as she gets closer, we see that it's a door. She reaches the door and opens it, stepping out into the bright sunlight. She looks around and sees that she's free. She turns around and looks back at the maze, now small and insignificant behind her. She raises her arms in triumph)
(JESSICA wakes up, smiling. We hear her voiceover again)
JESSICA (V.O.) I did it. I faced my fears and found my way out of the nightmare. I finally feel free.
(Closing shot of JESSICA standing at her window, looking out into the bright morning. The storm has cleared, and the sun is shining. She turns and walks back to her bed, ready to face whatever the day brings)
THE END
What is your reaction to this?
ral
I am sure that most readers would “approve” of this AI story as a good lesson about how to deal with nightmares, anxiety, etc. The algorithm has “learned” how we mostly think about such phenomena today and is reflecting it back to us. So it sounds like good or even sage advice. A new kind of narcissistic mirroring is taking place and we get giddy with the technological successes that reflect our own desires, values so very well. In thus being so enamoured with our own reflections through this mirror of AI, we are completely oblivious to the alien “horror” that lies just beneath this agreeable surface of reflection. You can get a sense of this alien presence if you encounter the “Blue Screen of Death” (Windows). Interesting that a death is required to open us to the presence of this alien reality.
The AI has no algorithmic access to marginal (and therefore challenging to the status quo) understandings of dreams, nightmares, or anything else for that matter. So on that score the AI remains silent and we are none the wiser. The possibility for a “new beginning” dries up!
All of these considerations are in terms of a technological simulation of the true state of affairs (i.e., psychic). We happily play about on the surface of life, getting giddy with excitement when something like an AI printout, news clipping, FB, Twitter gives us EXACTLY what we want to hear, enough to earn a “like”—whatever that means. Meanwhile an alien horror lurks just beneath the surface and is approaching the surface fast now. Too many signs to ignore yet that is precisely what we do—ignore them or choose the wrong target to blame for the rising tension. Yes, tension is the quality to focus on now. You can feel it everywhere but always with misleading interpretations. It has nothing to do with Trump or democracy now. It has to do with an alien reality, showing up as a horror in response to our shallow clinging to familiarity at all cost: “O yes, that is exactly the way to treat nightmares, get rid of them…” And the horror awaiting us deepens. Lastly, what I mean by “alien” is a reality that is alien to anything familiar. So our clinging to the familiar obscures any access to this reality.
Absolutely on the mark, John. I recall that the early psychoanalysts were called “alienists.” Perhaps it’s time for a new breed of alienists to make conscious what people do not want to see.
Absolutely on the mark, John. I recall that the early psychoanalysts were called “alienists.” Perhaps it’s time for a new breed of alienists to make conscious what people do not want to see.
As John W. says in his 1/12/23 reply to “Whose Fiction Is This Anyway?”, “The AI has nnoalgorithmic access to marginal understandings of dreams (or anything else). So on that score remains silent.” That is so true. AI is good for processing data & can easily learn dream symbols with their various meanings and construct average dream scenes, but what’s difficult for them is the unspoken, implicit knowledge — exceptions to the rules. Uncertainty. The AI’s screenplay of Jessica’s maze dream was written inside the MAZE of the SCIENTIST or Menpols. And, in that context — all AIs being fed data by the Menpols — the AI imposes a “normalized” moral framework for Jessica’s dream: Always go for the light. Flee the danger of the shadowy maze.
If Jessica were to Hump-like leap over the Menpols trapdoors and stay in the maze, she may well awaken her own imagination and discover all kinds of things, things that lurk apart from the mind that dreams them. Images and figures that are self-generating, ever-emanating entities with their own agenda. A Jung-like figure might appear to encourage Jessica to walk in the maze. To see where each twisting, turning corridor takes her. She might become aware of other realities in which she possesses all kinds of unusual, extraordinary powers. Look at the savants, multiple personalities in our world today, the power of hormones and psychedelics to change our personalities. Look at telepathy, clairvoyance, out-of-the-body experiences, projecting oneself to distant places, all of which have been documented. Can AIs do any of that? In moments of visionary intensity we experience how the apparent limitation of our powers is actually a limitation of our picture of ourselves. Is that where our muse is taking us? It’s definitely alien to our ordinary selves in the world presented to us by the scientists– the world we happen to be so”familiar” with — but must that alien-ness be perceived as horror? To some, no doubt, it will be, but to those who welcome and embrace the tsunami crashing down, or, as in a dream of mine, the boulder pulling you over the cliff?
As someone who has a background in both theoretical computer science (and complexity science) and depth psychology, I am very skeptical of any *general* claim that “AI has no algorithmic access to marginal understandings of dreams, nightmares, or anything else for that matter,” though it may be appropriate for how ChatGPT was trained (see paper “Training language models to follow instructions with human feedback” written by the team at OpenAI, the company which developed ChatGPT. The paper is on the OpenAI website and it provides additional details on how such technology is trained, but it is far from being comprehensive.). It seems to me that such AI models migh have (or develop) certain computational abilities to develop *rigorous* chains of analogical reasoning based on pattern recognition as well as access to a massive body of multidisciplinary work from which it could draw. It could then use these abilities to explore and amplify dreams in somewhat the same way as a depth psychotherapist might do. Obviously, the depth psychotherapist has advantages in this pursuit, but it seems to me that in many ways, AI could complement those advantages. For example, a depth psychotherapist might interpret a dream based, in part, on a fallacious analogy chain, one which does not hold up according to certain disciplines such as archeology, history, the textual expressions of the myths seen as being related to the dream, etc. But, AI could develop multidisciplinary chains (or even webs) of connections, chains which are formally and explicitly defensible as they draw from the best sources it used to form such chains. I feel that there may be a time where depth psychotherapists and AI work together, establish a meaningful dialog and division of labor, and together with their complementary skills work to deepen our understanding of the manifestations of the unconscious as well as to improve the rigor with which that is done so that it can both be defended and explicated to others.