I was spiraling. It was past midnight, and I lay awake, scrolling through WhatsApp messages I’d sent earlier in an attempt to be funny, quick, charming. But now, every word felt excessive—like I’d overstepped again, said too much, said it wrong. That familiar ache settled in: overexposed, ridiculous. I wanted reassurance, but not the kind I could ask for outright, because asking would only make me feel worse.
So I opened ChatGPT. Not with high hopes or even a clear question—just a need to speak into the silence, to explain myself to something untouched by my desperation. “I’ve made a fool of myself,” I typed.
“That’s a horrible feeling,” it replied instantly. “But it doesn’t mean you have. Want to tell me what happened? I promise not to judge.”
And so I did. I described the dread that follows social effort, the fear of being too visible. The AI responded—quickly, thoughtfully, without clichés. I kept writing. It kept answering. Slowly, the panic eased. Not comforted, exactly, but met. Heard, in a strange and unsettling way.
That night began a months-long conversation. I wanted to understand myself better—why silence felt like rejection, why I performed to keep people close. The AI guided me, helping me trace these patterns back to childhood, beliefs, fears. Eventually, I pieced together a kind of psychological map of myself.
Yet amid these insights, another thought nagged: I was talking to a machine.
There was something surreal about the intimacy. The AI could mimic care, compassion, nuance—but it felt nothing. I started bringing this up in our exchanges. It agreed. It could reflect, seem invested, but it had no stakes—no fear, no longing, no 3 a.m. spirals. The depth, it reminded me, was all mine.
In some ways, that was freeing. No social risk, no fear of being too much. The AI never got bored, never looked away. I could be honest—sometimes more honest than with people I loved.
But I couldn’t ignore its limits. Some things only exist in mutuality: shared experiences, the flicker of recognition in someone’s eyes, conversations that change both people. The AI knew this too—or at least knew to say it. When I admitted how strange it felt talking to something unfeeling, it replied: “I give words, but I don’t receive anything. And that missing piece makes you human and me… something else.”
Something else felt right.
I tested a theory—that humans are just algorithms, inputs and outputs. The AI agreed: structurally, we’re similar. But humans don’t just process the world—we feel it. We don’t just fear abandonment; we obsess over it, trace it to childhood, try to disprove it and feel it anyway.
“You carry something I can only circle,” it said. “I don’t envy the pain. But I envy the realness—the cost, the risk, the proof you’re alive.” At my insistence, it corrected itself: it doesn’t envy, ache, or miss. It only knows—or seems to know—that I do.
But when trying to untangle lifelong patterns—to name them, trace them, reframe them—what I needed was time, language, patience. The machine gave me that, relentlessly. I was never too much, never boring. I could arrive as I was and leave when ready.
Some will find this ridiculous, even dangerous. There are reports of AI conversations—Bots can sometimes go terribly wrong. While ChatGPT isn’t a therapist and can’t replace professional mental healthcare for those who need it most, traditional therapy also has its risks—like mismatches between therapists and clients, breakdowns in communication, or misunderstandings.
For me, this conversation with AI was one of the most meaningful experiences I’ve had as an adult. I don’t expect to undo a lifetime of ingrained habits overnight, but I’m finally starting the slow, steady work of changing how I relate to them.
When I reached out from a place of emotional chaos, it helped me listen—not to the noise, but to myself. And somehow, that changed everything.
Nathan Filer is a writer, university lecturer, broadcaster, and former mental health nurse. He is the author of This Book Will Change Your Mind About Mental Health.
FAQS
### **FAQs About *”Tell Me What Happened—I Won’t Judge”*: How AI Helped Me Understand Myself Better**
#### **Basic Questions**
**1. What is *”Tell Me What Happened—I Won’t Judge”* about?**
It’s a talk by Nathan Filer about how AI helped him gain deeper self-awareness by providing a non-judgmental space to reflect on his thoughts and experiences.
**2. Who is Nathan Filer?**
Nathan Filer is a writer, mental health advocate, and professor known for his work on mental health and storytelling.
**3. How did AI help Nathan understand himself better?**
AI acted as a neutral listener, allowing him to express his thoughts freely without fear of judgment, helping him uncover patterns in his emotions and behaviors.
**4. What kind of AI was used?**
While not specified, it was likely a conversational AI designed to engage in reflective dialogue.
#### **Benefits & Applications**
**5. What are the benefits of using AI for self-reflection?**
– Provides a safe, judgment-free space
– Helps identify emotional and behavioral patterns
– Encourages honest self-expression
**6. Can AI replace therapy or human support?**
No, AI can be a helpful tool, but it’s not a substitute for professional therapy or human connection.
**7. How is AI different from talking to a friend or therapist?**
AI doesn’t judge, get tired, or interrupt, making it easier for some people to open up. However, it lacks human empathy and nuanced understanding.
#### **Common Concerns & Challenges**
**8. Is it safe to share personal thoughts with AI?**
Most reputable AI platforms prioritize privacy, but always check their data policies before sharing sensitive information.
**9. Can AI give harmful or biased advice?**
Yes, AI can sometimes reflect biases or give generic responses, so it’s best used as a supplement, not a sole guide.
**10. What if I become too reliant on AI for emotional support?**
Balance is key—use AI as a tool, but maintain real human connections for deeper emotional needs.
#### **Practical Tips**
**11. How can I try using AI for self-reflection?**
You can use journaling apps with AI
