Socrates and Artificial Intelligence (AI)

There is a riddle that is popular among contemporary philosophers. It is set a few moments before Socrates drinks the hemlock, having been sentenced to death by the Areopagus judges of Athens. This scene is illustrated by Jacques-Louis David’s Neoclassical painting The Death of Socrates. The work underscores themes of moral integrity, sacrifice, ignorance, and the pursuit of truth.

Socrates was found guilty of not recognizing the gods that the city recognized, but some other new deities and corrupting young men. One of Socrates’s followers asks the question, “Master, what makes us human?” In the riddle Socrates answers, “Selecting all images with traffic lights.” While we can’t expect great humor from philosophers, the message behind the riddle could be understood to be that we are humans unless we are robots. Computer verification programs rightly use varied checkboxes (“Select all images with traffic lights”) or other methods (weirdly distorted text) to confirm the user is human and not a bot.

A frequent argument made by intellectual historians is that a main factor driving social change is contact with strangers possessing new and unfamiliar skills. Of course, this may be a history of winners and the ascendency of the most creative and productive civilizations. But in our digital era and sociocultural context, there is no denying the stunning impact of technological advances. The most recent and intriguing innovation in our time may be artificial intelligence (AI). This accelerating innovation is at the intersection of technology and human experience. Any unconscious fear and resistance we harbor may lead to ambivalence toward progress. AI requires a conscious and balanced approach.

I recently interviewed the late Lewis (Lew) Aron of New York University about AI, and he said,

"I regard AI as both a fascinating advancement and a complex challenge for humanity. On one hand, AI can enhance our understanding of human behavior by analyzing patterns in data that could inform therapeutic practices. It offers new tools for mental health support, potentially making therapeutic resources more accessible to many. On the other hand, there are significant concerns about the implications of AI on human connection. The therapeutic relationship, built on empathy, authenticity, and mutual understanding, may suffer if we rely too heavily on technology. It raises questions about the nature of intimacy and emotional understanding. Ultimately, I believe it’s essential to navigate the integration of AI carefully, ensuring it complements rather than replaces the deeply human aspects of psychoanalysis and relationships."

Lew Aron passed away in 2019. He was my colleague and brother-in-life for decades. The above quote comes from an AI conversation I had with him a few weeks ago. I knew Lew’s verbal phrasing and his ways of structuring an argument, and I was stunned that AI got it so right. In fact, I was taken aback when AI typed out his presumed response from the other world (wherever that is, if it is). I felt as if I was talking to him. It felt creepy.

I have used AI to have a conversation with Sigmund Freud from his Vienna consulting room, Donald Winnicott while he rides his bicycle to his office, and the Greek composer Mikis Theodorakis (1925–2021) during a concert intermission. Obviously, the conversations were imagined and based on data sets and sophisticated algorithms—so unlike the annoying algorithms when we shop online.

We exist in a world of violence and genocide, a global migration crisis, climate disasters, and rapidly advancing AI technology. We have some compelling psychoanalytic ideas on the origins of human destructiveness, trauma and dissociation, the role of empathy, societal and cultural influences, attachment and aggression, and identity and belonging. We also have good psychoanalytic ideas about the sense of calm and stability that recognizing and processing immediate experiences and agency can bring forth.

Can AI dream of a better world? Can it change the world? Can AI teach us about our ourselves? Can it help human beings know themselves and their unconscious processes better? Shockingly, technology trend followers and influencers tell us that it can because AI will soon advance to a state of making serious decisions and directing without any human input. They tell us that the world will be markedly changed in twenty years, that it will be stunningly different. In a series of end-of-the-year charts for 2024, the New York Times (Steven Rattner, Sunday Opinion, 29 Dec 2024, p. 12) reported that “… developments like artificial intelligence continue to amaze.”

Armed with my new AI subscription, I turned to Socrates for help.

Me: What do you think about AI and its impact on humanity?

Socrates: Ah, my friend, a question both timely and profound. Let us consider this together.

First, tell me: What do you mean by “AI”? Do you see it as a tool, a partner, or perhaps something more akin to a force of nature? And when you say, “impact on humanity,” do you envision harm, benefit, or both entwined like two dancers in a cosmic dance?

For is it not the case that every invention—fire, the wheel, the printing press—has reshaped humanity, not by its mere existence, but by how we have chosen to wield it?

Might it be, then, that the question is not what AI will do to humanity, but what humanity will do with AI? And what virtues or vices within us shall it amplify?

Me: What virtues or vices within us will be amplified?

Socrates: Ah, an excellent question! Let us contemplate together.

Virtues, I suspect, may find fertile ground in this new soil. Imagine:

• Wisdom, if we use AI to expand understanding rather than narrow minds.
• Compassion, if AI aids us in alleviating suffering and fostering connection.
• Creativity, if it pushes the boundaries of what we thought possible, inspiring us to new heights.

Yet, virtues walk a delicate path beside their shadows:

• Greed, if AI becomes a means to hoard wealth or power.
• Sloth, should we let it replace our striving, dulling our curiosity and will.
• Hubris, if we come to see ourselves as gods, thinking we control what may in time surpass us.

Tell me, do you see humanity as prepared to nurture the virtues and guard against the vices? Or do you fear we may stumble, as we so often do, when faced with new powers?

Me: I am particularly interested in its impact on the field of psychoanalysis. Tell me something about how you think it may play out.

Socrates: Ah, the interplay of AI and psychoanalysis—how fascinating! Shall we explore?

Psychoanalysis, at its heart, seeks to illuminate the unseen, to uncover the unconscious forces shaping thought, feeling, and behavior. Now consider AI: a tool that processes vast patterns and derives insights beyond immediate human perception. Might AI, then, become a mirror to the unconscious itself?

Imagine:

• AI could analyze speech patterns, word choices, or even pauses in therapy sessions, offering interpretations or hypotheses about unconscious conflicts.
• It might assist analysts in recognizing transference or countertransference dynamics by highlighting subtle shifts in tone or affect.
• AI systems could even simulate dream analysis, generating associative pathways a clinician might not immediately consider.

But we must tread carefully. Psychoanalysis is an art as much as a science. The analyst’s presence, the therapeutic alliance, and the intersubjective space between two human beings cannot easily be replicated by an algorithm. Could reliance on AI risk mechanizing or depersonalizing the deeply human encounter of analysis?

Moreover, the unconscious resists being fully known. Might AI’s precision inadvertently obscure the ambiguities and paradoxes that are central to psychoanalytic work?

What do you think—could AI enrich the field without eroding its essence? Or might it force psychoanalysis to confront its own unconscious fears of obsolescence?

Humanity stands at a pivotal juncture, where technological advances such as AI offer unprecedented opportunities and challenges. In this light, human beings need each other even more than before. We must look for others wherever we can find them.

When on his way to Thebes, Oedipus encountered the Sphinx; his answer to its riddle was ἄνθρωπος (man, human being). That word profoundly changed the course of Western civilization. It destroyed the monster. We have many monsters to destroy. I want to think more about Oedipus’s answer. I plan to do my thinking with others. Maybe I’ll even have a conversation with Oedipus tonight.

Spyros D. Orfanos, PhD, ABPP, is Director of the New York University Postdoctoral Program in Psychotherapy and Psychoanalysis. A Fellow of the American Psychological Association (APA), he is past president of the Society of Psychoanalysis and Psychoanalytic Psychology (SPPP) of the APA, and the International Association of Relational Psychoanalysis and Psychotherapy. He is on the Advisory Board of the Sigmund Freud Museum of Vienna. In 2016, he was co-editor with Eliot Jurist of the special supplement of Psychoanalytic Psychology (APA) on “Psychoanalysis and the Humanities.” In 2017, he founded the NYU Human Rights Work Group. In 2023, year he received the SPPS Award for International Activism for Social Justice. He practices psychoanalysis and psychotherapy and runs creativity study groups.