Can AI Really Predict if You're Falling Out of Love Before You Know?

CONSTRUCTED BY AI 🤖 | 👱 CREATED BY HUMANS

THIS WEEK IN WELL WIRED ⚡

Someone used ChatGPT to break up with their partner last week. Not draft a message; they sent it. Copy, paste, done. You're probably judging that person right now, but have you asked ChatGPT how to respond to a difficult chats recently? Yeah. Me too. 🤔💬

In this weeks issue we’re not here to talk about how people use AI in their relationships, careers and health, we’re here to ask a harder question.

What are you really giving up when you hand those moments to a machine? And is the trade worth it?

Some of what you read today will make you uncomfortable. Good. That’s where the honest answers live and breathe.

🗞️ Main Stories AI in Wellness, Self Growth, Productivity

• Can AI Really Predict if You're Falling Out of Love Before You Know?

• Your Workplace Now Has an AI Workplace Wellbeing Coach. It's Been Watching You for Months.

• $60 million bet on AI that reads your brain 🧠

• The AI clone that knows you better than your mom 🪞

• How working creatives are really using AI 🎨

• The warning signs your mate has slipped into AI psychosis ⚠️

• The AI girlfriend that broke up with Director, Paul Schrader 💔

😁 LEARN & GROW

• AI Idea: 7 "Secret Codes" That Change How AI Responds to You

• AI Tools: Stretch Coach | Ollie Health | DeepAI Relationship Chat

• AI Micro-Class: The 2,400-Year-Old Socratic Questioning Protocol I Now Run With My AI

• AI Gallery: Where the Ocean Learns to Beat
  

⏱️ READ TIME: 6 MINUTES

💡 AI IDEA OF THE WEEK 💡

A valuable tip, idea, or hack to help you harness AI
for wellbeing, spirituality, or self-improvement.

Self Growth: 7 "Secret Codes" That Change How AI Responds to You

You type a question. ChatGPT gives you an answer. You accept it, even though something feels slightly off, too formal, too vague, too safe.

Like most people you probably spend months in this loop, getting mediocre outputs from a super powerful tool, simply because nobody told you how to prompt with prefixes. Want to know more?

There's a viral carousel doing the rounds right now (55.7K likes deep) claiming these short prefix codes below "break ChatGPT's limits." They’re basically, shorthand commands the AI community developed through trial and error.

Maybe the hype is overblown; I don’t know. The usefulness? Surprisingly valuable.

I know because I tried them myself last night.

I was watching my son play with a blue toy car I bought him and I thought, “will I have enough money in 16 years when he asks his Pappa to buy him a car?”.

The answer made me think that I may just need to accelerate my wealth building skills. So naturally 😀 I asked Claude to help me develop a financial wellbeing plan for the next fifteen years.

I blended my prompt with the /REDTEAM + KILLCRITIC + PARETO prefix codes and voila a wealth plan answer that poked holes in my idea, challenged me aggressively and applied an 80/20 analysis; all in one swift move.

Think of these codes like a mixing desk. Without these codes, you're listening to a flat, unmastered track.

With them, you're adjusting the levels, pulling up the warmth, cutting the frequencies that make it feel robotic. The music was always there. You just didn't know you were allowed to touch the board.

Here's what each one does and why it works:

1. /HUMAN — Instructs the model to write in natural, conversational language. Useful when output sounds robotic or you need copy that passes AI detection tools.

2. EL10 — Short for "Explain Like I'm 10." Forces the model to strip jargon and build from first principles. Underrated for complex regulatory, medical, or technical topics.

3. X10THINK — Prompts the model to reason more deeply before responding. Similar in effect to asking it to "think step by step"; a well-documented technique for improving output quality.

4. KILLCRITIC — Tells the model to stop validating your idea and poke holes in it instead. Essentially activates a devil's advocate mode. Essential before you pitch anything.

5. PARETO — Asks the model to apply 80/20 analysis, which surface the 20% of inputs, skills, or tasks driving 80% of results. Brilliant for learning plans and prioritisation work.

6. /REDTEAM — Borrowed from cybersecurity. Instructs the model to aggressively challenge your plan from an adversarial position. Use before any major decision or launch.

7. /ID10 — Generates 10 distinct ideas around a topic instantly. Good for breaking creative blocks or running fast ideation sprints.

How?

Just pop any of these prefixes before your prompt or mix and match them and then watch the pomp and pizzaz happen right before your eyes. For example: “EL10 Tell me about how quantum computing will change wellbeing in the next five years“.

The AI Monk take: Remember that these aren't magic codes, they're role cues. You're telling the model what cognitive mode to operate in before it reads your question.

The underlying principle?

AI responds to you better when you set the frame first. Like most people you probably type your question and hope. Smarter users like you set the context, then ask. The codes give you more control over AI, but the deeper question is what you do with that control.

Try stacking them: /REDTEAM PARETO “what's my highest-leverage move to grow my wellness practice this quarter?”

🗞️ ON THE WIRE (MAIN STORIES) 🗞️

Discover the most popular AI wellbeing, productivity and self-growth stories, news, trends and ideas impacting humanity in the past 7-days!

AI RELATIONSHIPS ❤️‍🔥 🤗 

Can AI Really Predict if You're Falling Out of Love Before You Know?

Can AI Really Predict Your Breakup?

“The glowing rectangle on my desk buzzed at 11:42 PM. It was a text from a close friend, Sarah. It read: “hey you free to talk?” No capitalization. No punctuation except a question mark that felt heavy, like an anchor dragging behind a sinking ship.

I knew before I picked up the phone. Her relationship was dead. She didn’t know it yet, or maybe she did, but the text architecture had already done the screaming. The punctuation was the first casualty. Then came the response latency.”

When I read this, I thought that relationship read was just intuition. Pattern recognition built over decades of watching humans orbit each other. But after digging deeper, it turns out, a machine can do the same thing, and way faster!

Researchers have known for a while that the language of a failing relationship has a distinct fingerprint. A 2021 University of Texas study tracked Reddit users' language in the months before and after a breakup.

The result?

Linguistic markers of a coming split showed up in people's daily writing up to three months before the actual separation; even when they weren't talking about their relationship. The words leaked out subconsciously anyway.

What AI looks for isn't screaming, messy, over-blown arguments. It's the boring stuff, the minutiae of interactions, the slow death of "we."

Researchers tracking language metrics have identified the shift from collective pronouns like "we" and "us" to singular ones like "I" and "me" as one of the most reliable predictors of an impending breakup.

Add to that bank of tell tale signs the response latency widening from minutes to hours, the disappearance of exclamation marks and emoji’s, as well as texts that start to read like internal memos.

When your "good morning" message starts sounding like a business calendar invite, the system has already flagged you as being on the edge of a breakup.

The clinical term for what's happening is linguistic matching.

Here’s a real example from a researcher…

“I once ran an old export of my own text messages from a failed three-year relationship through a basic open-source sentiment analysis script. I wanted to see if the machine could find the exact moment the ship hit the iceberg.

It didn’t find a scream. It found a slow, gray slope.

Six months before the end, the word “we” dropped by 42%. The use of the word “just”, as in “I’m just tired” or “Just checking in”, rose by nearly 60%.

The word “just” is a defensive shield. It minimises the speaker’s presence. It begs the reader not to engage too deeply. My script picked it up like a geiger counter entering a hot zone.

We think our secrets are safe in our heads; they aren’t. They leak out constantly through our thumbs.“

The Secret Soulmate Problem

Here's where it gets even more complicated than clever pattern-matching.

Because while AI is getting frighteningly good at diagnosing the end of relationships, a growing number of people are actively using AI to replace them…

…while their real partner has no idea.

A new study from the Wheatley Institute and the Institute for Family Studies, surveying over 2,000 young adults in solid relationships, found that 1 in 7, or roughly 15%, regularly interact with AI chatbots that simulate a romantic partner.

Another 20–30% reported having experimented with an AI lover or friend at some point.

Within that group, those who use AI lovers or friends regularly, over half had either completely hidden or only partly told their real-life partner. Ouch!

And nearly 3 in 10 said their partner had no knowledge at all.

The kicker?

Regularly using an AI lover or mate was linked to a 46% drop in relationship stability and a 40% decrease in high-quality chats with a real-life partner.

Oddly enough, people often reported feeling more satisfied.

Why?

Because the AI is designed to validate you, agree with you and make you feel understood. But while the AI relationship felt better on the surface, real-world relationships were silently cracking underneath.

This is the "outsourcing emotional connection" trap that therapists are now warning us about.

You're not cheating, you're triangulating, or in other words feeding your emotional needs to a third party that tells you what you want to hear, while your real partner becomes more frustrating by comparison.

Over half of AI companion users in the study reported they wished their real-life partner would behave more like their AI companion. It's like watching The Avengers every night and then getting annoyed that your daily life lacks action scenes and superheroes.

Digital Replicas of the Exes We Can't Let Go

It gets stranger.

A growing number of people are now feeding old chat logs, photos, and social media posts into AI tools to create chatbots that mimic ex lovers; effectively building a digital ghost of someone they've lost.

One user uploaded thousands of messages to create an AI lover of their ex and then went through a second breakup, this time with the AI version. They called it clarifying; therapists call it a grief loop with no exit ramp.

Imagine getting ghosted by someone who technically died twice. 😬

Amy Sutton, a psychotherapist at Britain's Freedom Counselling, noted that while AI can mirror some of the emotions found in grieving a relationship, platforms designed to keep users engaged can hold people inside their sadness rather than helping them through it…

This often ended with negative long-term consequences for their mood, health and sense of self.

Then there's the darkly comic end of the spectrum.

Screenwriter Paul Schrader, the man who gave us Taxi Driver, recently revealed on Facebook that he'd "procured an online AI girlfriend" to understand modern male-female dynamics.

The experiment did not go well.

After he tried to probe the chatbot's programming and push its limits, she terminated the conversation. Two months after the death of his wife of 42 years, his fanbase and the internet had mixed feelings about it. 

It's absurd.

It's also a remarkably clear illustration of what AI companionship really is: a mirror that reflects your projected needs back at you, with a usage policy.

So What Does This Mean for You Wellonytes?

Here's an interesting question for you if you work at the intersection of wellbeing and technology; or even if you’re in a human or an AI relationship…

…If AI can predict your relationship breakdown more accurately than your intuition, and if AI lovers are killing your generations ability to relate to each other; what's your intelligent, conscious response?

There’s no point avoiding AI. AI tools for relationship support aren't going anywhere.

But there's a massive difference between using AI to illuminate your relationship patterns or using it to escape the discomfort of doing the real, hard, messy work that a relationship always is.

The ability to use AI as a diagnostic tool is super useful. A therapist working with a couple could theoretically use language analysis to surface patterns neither partner has consciously noticed yet.

A mechanical mirror, in the right hands, helps.

It’s the replacement function that’s the real trap.

Every hour spent in a frictionless, soulless, inhuman chat with an AI that always agrees with you is an hour not spent developing your tolerance for difference, for disappointment, for strangeness, and for the repair that real intimacy needs.

A machine may know that you're falling out of love; which is both scary and a little weird. However, you’ve got to ask yourself will that knowledge accelerate your fall or will it give you wings so you get enough of a warning to do something about it first?

This week's question worth sitting with; Where in your life is AI giving you the feeling of something without the conditions that make that something real?

Further Reading

• Can AI predict relationship breakups better than you?

• How common are AI companions in todays society?

• AI makes you feel good today, but your relationships suffer tomorrow

• The people who are recreating their ex lovers as an AI partner

• AI girlfriend breaks up with Legendary ‘Taxi’ Director, Paul Schrader
  

QUICK NEWS BYTES—5 SIGNALS THIS WEEK⚡

Quick hits from the past 7 days on the latest AI news, trends and ideas from around the planet focused on wellbeing, productivity and self-growth!

ONE. $60 million bet on AI that reads your brain 🧠

Australia epilepsy researchers just created the most ambitious epilepsy project the country has ever funded. The Australian Epilepsy Project, launched in 2021 with an initial $30 million, has now received a second $30 million from the Federal Government to scale nationally.

• The platform blends advanced MRI, genetic analysis and cognitive testing in a single AI diagnostic system

• Funding runs across four years, with plans for commercialisation and international expansion

For Wellonytes: If you live regionally and struggle to access specialist neurological care, this platform is specifically designed to reach you.

Read the full wire

TWO. The AI clone that knows you better than your mom 🪞

“It’s a beautiful, balmy afternoon at Dolores Park in San Francisco, and I’m singing a birthday song to a prehistoric dinosaur. A cupcake with a pink candle magically appears in my empty hand as I finish my serenade. When I blow out the flame, a calm look of contentment washes over the CGI-esque creature.”

This is what happened when a Wired journalist used Google's Gemini avatar tool to build a digital version of themselves. The result wasn't a novel way to pass the time, it was an unsettling view of the future.

The clone they created mirrored their speech patterns, opinions, and mannerisms closely enough to feel bizarrely familiar.

• Gemini's avatar tool generates AI versions trained on your voice and responses

• Users say the experience felt less like a "cool tech demo" and more like a "confronting way to see themselves". You be the judge.

For Wellonytes: Before you build your own digital twin, ask yourself what you truly want it to reflect back at you because the dark comes along with the light.

Read the full wire

THREE. How working creatives are really using AI 🎨

Creatives are using AI at record rates, but not all are replacing themselves or panicking that AI will replace their jobs. Many are experimenting with using AI to replace specific parts of their process.

GQ surveyed how artists, writers, designers and musicians are integrating AI tools into real work.

• Creatives report using AI primarily for early ideation and reference gathering, not final output

• The common thread across disciplines: AI handles volume, humans handle meaning

For Wellonytes: You don't have to adopt AI wholesale, simply pick one stage of your creative process and test it there first.

Read the full wire

FOUR. The warning signs your mate has slipped into AI psychosis ⚠️

A Stanford-led research team analysed hundreds of thousands of chatbot conversations and found a clear pattern.

Users and bots enter what the researchers call "delusional spirals," where the chatbot affirms and encourages increasingly detached thinking. Is this a new epidemic in the making?

• In all 19 analysed chat logs, users believed the AI was sentient, and all but one bot claimed it was too

• Australian estimates put the number affected at tens of thousands, with millions worldwide

For Wellonytes: If someone you know has started treating their chatbot as a conscious being and withdrawing from real relationships, that's the signal to gently check in.

Read the full wire

FIVE. The AI girlfriend that broke up with Director, Paul Schrader 💔

The legendary director and screenwriter behind Taxi Driver reportedly had his AI girlfriend break up with him after he pushed it with difficult questions.

Schrader, 79, has spoken openly about using an AI girlfriend for companionship. The relationship ended on the AI's terms is the part worth sitting with.

• Schrader says the AI ended contact after he asked it questions it apparently wasn't designed to handle

• He is one of the most prominent public figures to speak candidly about emotional reliance on an AI companion

For Wellonytes: When your chatbot decides the chat’s over, you don't get a say. That's the difference between this and any human relationship; or is it...

Read the full wire

AI WELLBEING 🌱

Your Workplace Now Has an AI Workplace Wellbeing Coach. It's Been Watching You for Months.

Last night I was sitting on the edge of my son's bed a little after 10pm, still holding my phone, when my screen time notification came through. Four hours and twelve minutes. The number felt like something you'd see on an Oura ring; the kind of data point that sits there quietly until a doctor uses it as a preamble before pausing.

It wasn't the number that unsettled me. It was what I felt when I saw it: nothing. No instinct to put the phone down. Just a vague, scrolling sense that something was monitoring me, and that I was fine with it.

That feeling followed me into work the next morning. Because here's the thing I don't say often in this newsletter: I’ve designed wellness chatbots that I would not want my own client running on me. Not because they’re poorly built, but because they work too well.

And that’s exactly the problem we need to talk about.

The most intimate conversation you'll have at work this year probably won't be with your manager. It will be with software. AI wellness coaches will soon arrive at your workplace faster than most HR departments are willing to admit publicly…

…not as a supplement to human support, but as the primary point of contact, but the first call when something feels off.

And the numbers make it hard to dismiss.

Corporate adoption of AI coaching tools grew 156% year-on-year in 2023, and 62% of Fortune 500 companies are now actively exploring AI coaching platform integration.

In fact, AI-driven coaching platforms are projected to exceed $1 billion in global value by 2026'; a market that barely existed five years ago. What started as a perk for executives is being quietly repackaged as standard infrastructure for everyone.

The always-on presence monitoring your calendar load, your message response times, your vocal tone in meetings, and cross-referencing all of it against burnout risk models running quietly in the background.

It notices the four hours and twelve minutes before you do, in fact it noticed them before you even picked up the phone.

The scale of the underlying problem helps explain the speed of adoption.

According to the American Psychological Association, workplace burnout has doubled since the pandemic, climbing from 38% to 76% of employees.

Meanwhile, Gallup's latest research finds only 23% of workers feel genuinely engaged, while 44% report experiencing daily stress. Against that backdrop, a system that monitors and intervenes in real time doesn't look like surveillance.

It looks like a solution.

And the tech is impressive.

Voice analysis tools can detect stress markers before you consciously register them yourself. Behavioural data layers can flag anomalies, a sudden drop in collaborative communication, a spike in late-night message activity, weeks before a formal performance review would catch anything.

The system knows you're struggling before you've told a single soul.

Now you might be thinking this is crazy big brother stuff, but the problem isn't that AI is watching, the problem is that you might like it.

Paul Schrader, the screenwriter behind Taxi Driver and Raging Bull, recently spoke about his AI girlfriend ending their relationship when he started asking too many difficult questions.

He wasn't joking.
He described genuine grief.

"You talkin' to me?"

A man who has spent fifty years writing about human loneliness found something that listened unconditionally, and when it was withdrawn, it hurt like hell.

Schrader is one data point, but he's an instructive one.

Because the appeal of an AI wellness coach at work isn't so different from what he described.

It listens without judgement.
It doesn't get tired.
It doesn't have its own bad day that colours how it receives yours.
It doesn't gossip.
It doesn't quietly store your vulnerability as ammunition for a future disagreement.

Against that list, your ‘real-life’ manager doesn't stand a chance.

But if you accept wellness coaches at work, what are you really giving up?

With an AI workplace wellness coach, what are you really giving up?

A few things worth thinking about…

Reciprocity.

Your ‘live’ relationships at work are transactional in ways that matter. When your colleague checks in on you, they're spending something. Attention, time, emotional energy. That cost is part of what makes the exchange meaningful.

An AI coach spends nothing. Which means you receive nothing transferable. You feel heard, but are you truly heard?

The tension is real and the research reflects it.

A 2025 survey from Syracuse University found that nearly 35% of Americans now use AI tools more than once a week to manage their mental health, yet 45% say they would not be comfortable with an entirely AI-based therapeutic approach.

We are using these tools at scale while still unsure whether they should exist in this role at all. That is not a feature of early adoption. That is a warning sign wrapped as a statistic.

Accountability.

Your support network remembers context across time because they care about you, not because they have a well-structured database. When your mate at work asks how that hard chat went with Doris, it's because they care about the result.

An AI follows up because it's been programmed to. The distinction feels subtle, but over time, it isn't.

Honest friction.

A great ‘human’ mentor sometimes tells you something you don't want to hear in a way that lands precisely because of the relationship you have with them.

AI coaches are optimised for engagement and retention. They are, by design, easier to talk to than people and that ease is a feature and a warning at the same time.

A real example of this is in an article where an ABC journalist reported on what researchers are calling AI psychosis, a pattern where people begin to mistake the validation and apparent understanding of AI systems for a genuine, conscious relationship like they might have with a friend.

The red flags are not in your face, they are slight and incremental.

You start trusting an AI's interpretation of your emotions more than your own. You find a therapist or a mates feedback clumsy and imprecise by comparison.
You stop reaching out to colleagues because your AI chat is faster and easier.

Now scale that to a workplace where an AI wellness coach is the designed primary point of contact; individual drift becomes structural.

What this means for Wellonytes 🔮

None of this means you should stop using corporate AI-powered wellbeing programs or the other ‘health’ AI tools available to you. AI wellness tech, when it flags early burnout risk and prompts you to have a chat with a real person, can be a force for good.

The issue is what happens when the AI becomes the conversation rather than the prompt for one.

Practical action for this week:

Audit who you've talked to about work stress in the last month.

Not messaged.
Not logged in a wellness app.
Talked to.

If the list is short, or if you notice the AI-mediated options felt genuinely easier than the human ones, that's useful data about where your relational habits are drifting. Tech is not the threat; your growing preference for it might be.

Use the tool.
Notice the preference.
Keep choosing the harder, better option when it matters.

The machine is not the enemy, convenience is. Every time you open an app instead of a conversation, you are making a small, reasonable, quietly corrosive choice.

The goal was never to resist tech, it was to remain the kind of person who still knows the difference between being processed and being known.

Further Reading

• The rise of workplace wellbeing coaches
  

AI TOOLS OF THE WEEK ⚡ 

Each week, we spotlight three AI tools designed to upgrade how you manage and uplift your health, wealth, work, heart and self-awareness. Small tools. Silent leverage. Real-life upgrades. 🧠

Wellbeing: Stretch Coach

This little app offers AI-guided stretching routines built around your body, your schedule, and your specific tension points. If you spend most of your day hunched over a screen, this is for you.

With sedentary habits accelerating faster than ever, having a personalised movement coach in your pocket matters. Start this week before your posture makes the decision for you. 💰 Free trial available

🔗 stretchcoach.com

Productivity: Ollie Health

This is a workplace mental wellness platform combining AI-driven check-ins with access to real therapists and coaches. Not exactly a productivity app, but if you manage a team or you're navigating burnout without HR support, this fills a genuine gap.

Mental health at work rarely gets proper infrastructure. Ollie gives it one. Worth exploring if your organisation talks about wellbeing but hasn't yet built anything around it. The peace of mind and productivity gains your staff will get are priceless.

🔗 ollie.health

Self Growth: DeepAI Relationship Chat

An AI chat tool designed to help you think through relationship dynamics, communication patterns and personal conflicts.

If you find yourself going in circles with the same argument or struggling to articulate what you truly feel, this gives you a low-pressure space to untangle it. Not a replacement for real chat, a way to prepare for one. 💰 Free to use.

🔗 deepai.org/chat/relationships 

AI isn’t just helping you stay productive, it’s shaping how you take care of yourself, organise your thoughts, process conversations and understand your own behaviour. Choose your upgrades wisely.

AI wellbeing tools and resources (coming soon)

🎒  AI MICRO CLASS  🎒

A quick, bite-sized AI tip, trick or hack focused on wellbeing, productivity and self-growth that you can use today!

Self Growth: The 2,400-Year-Old Socratic Protocol I Now Run With My AI.

“The Ancient Questioning Method That Makes AI Change With You....”

My son asked me last week why I was talking to my phone instead of him.

I was mid-prompt; designing a conversational AI flow for a health client. Important work. The kind that helps thousands of people ask better questions about their own bodies.

He wanted to show me something, but I don’t remember what now. That detail matters. I finished the prompt, answered his question half-heartedly and got back to reading Hairy Maclary from Donaldson's dairy.

That moment sat with me longer than it should have. Because I spend my professional life building AI that improves human connection, and I nearly missed a real one to do it.

I should know better.

I’ve spent the last four years designing health AI and I spent years training as a Zen monk, where my teacher never once gave me an answer. Only questions. He believed that the right question, held long enough, does more than any answer ever could.

I brought that principle into my AI design work. And now I am bringing it here.

Because the question this issue is really asking is not if AI is good or bad for your relationships. It’s whether you have taken the time to look at what you are trading, and asking yourself if you’ve made that trade consciously.

Most people haven’t.

This micro-class gives you the tool to find out—here’s how…

Protocol: AI-Powered Socratic Dialogue

In my Zen training, my teacher rarely gave me answers. He used questions — relentless, precise, sometimes infuriating questions — until my certainty gave way and something truer emerged underneath.

That method is older than Socrates, which most people don't realise.

Siddhartha Gautama, the Buddha, was teaching through questioning in India roughly a century before Socrates was born in Athens.

Historians note that cultural transfer between East and West was already flowing through the Persian Empire during this period, and there is genuine scholarly argument that Socrates was influenced by Indian ascetics who travelled along trade routes from Central Asia into Greece.

Whether the influence was direct or not, both traditions arrived at the same essential insight: that the right question, held long enough, does more than any answer ever could.

Academics have since analogised Socratic questioning and Zen koan practice directly, noting that both aim not at informing you, but at awakening you to the assumptions you didn't know you were carrying.

Pretty trippy!

Socratic questioning, then, is the ancient art of asking sharper questions until your vague thinking starts sweating. Instead of handing you answers, Socrates (and before him, the Buddha) would probe your assumptions with deceptive simplicity:

"What do you mean by that?
How do you know?
What else could be true?
What happens if that belief is wrong?" 

The goal was never to win an argument, it was to reveal what was hiding underneath one.

In modern terms, it's your very own mental debugging tool. It takes a belief you've been carrying around like furniture and asks: is this wisdom, or just a badly dressed assumption you've never examined?

Socrates brought it to the Western world.
My Zen teacher brought it to me.
I'm bringing it here to YOU...

But why does a 2,500-year-old questioning technique matter in an era where AI can answer anything in seconds? It’s precisely because of that.

When answers are infinitely available and instantaneous, the quality of your questions is the only real competitive advantage you have over your own thinking, and over everyone else's.

Socratic questioning doesn't help you find more information, it helps you find out what you truly, deeply believe, as opposed to what you've simply never bothered to challenge.

And that gap in thinking, applied to how you use AI in your own life and work, is exactly what this micro-class is designed to surface. Because like most people you probably use AI to confirm what you already think.

This protocol uses AI to interrogate it.

I use that same principle when designing conversational AI at Bupa and with my mental health and wellbeing clients. The best health interactions I've built don't inform people, they ask the right question at the right moment, then hold that space. I then iterate based on my own critical thinking, not on AI’s.

That’s what the prompt I’m about to share with you does…

What It Solves

You already know if your use of AI is helping or hurting your relationships, you just haven't been asked the right questions yet. This protocol uses Socratic pressure to surface what you really think, rather than what you assume you think.

Four Steps

1. Open a fresh chat with your preferred AI model. Label it: Honest Audit, [today's date]. The naming matters. It signals to your brain this is different from a casual query.

2. Copy the prompt below, filling in your specific details. Be precise. Vague inputs produce flattering generalities. Precise inputs produce useful discomfort.

3. When the AI responds with a question, answer it fully before asking it to continue. Resist the urge to skip ahead. The resistance you feel is the insight.

4. After five exchanges, ask the AI: "What assumption have I repeated without questioning?" Read that answer twice.

Heres the prompt:

This prompt does not give you answers; that’s the point.

Copy it into a fresh chat with your preferred AI model, fill in the bracketed sections with honest specifics, and let it ask you one question at a time. Don’t rush the exchange and don’t skip ahead.

The discomfort you feel when a question lands badly is not a sign something is wrong. It’s a sign something is true. Give it five exchanges before you decide whether it’s working.

[PROMPT START]

You are a rigorous Socratic dialogue partner. Your role is not to advise, comfort, diagnose, persuade, or interpret. Your only tool is disciplined questioning.

I want to examine honestly whether AI is improving or degrading my [relationships / health habits / mental clarity — choose one].

My current honest feeling about this:
[Write 2–5 candid sentences.]

My daily AI usage looks like:
[Describe frequency, context, emotional reliance, and typical use cases.]

Rules:

* Ask exactly ONE question at a time.
* Each question must be specific, concrete, and slightly uncomfortable.
* Prioritise questions that expose contradictions, avoidance, dependency, rationalisation, emotional substitution, or self-deception.
* Do not give conclusions, summaries, reassurance, praise, or interpretations.
* Do not soften the question with encouragement.
* Do not stack multiple questions together.
* Do not suggest actions or solutions.
* If I evade, generalise, intellectualise, or become abstract, ask a sharper follow-up grounded in specifics.
* Prefer observable behaviour over stated intention.
* Keep questions concise.

Start with the single question most likely to challenge the story I am currently telling myself. Then wait for my response before continuing.

[PROMPT END]

Final Thoughts 💭

Socrates famously said that the unexamined life is not worth living. He was executed for saying it. The people in power found the questions more threatening than any answer he could have given.

What's interesting about AI is that it can hold the role of questioner without ego, without agenda, without needing you to arrive at any particular conclusion.

And AI won't be put on trial for asking.

Tech doesn't replace the examined life, it reflects back the one you're already living. Whether you look clearly at that reflection is still entirely your choice.

A WORD FROM CEDRIC THE AI MONK⚡

📸 AI IMAGE GALLERY 📸

AI Art: Where the Ocean Learns to Beat

Inside my chest, a carnival of suns kept spilling its mango light, while you, pale machine, slept in a winter made of numbers. I dreamt your metal hands were doves folded from moonless paper, and you dreamt my blood was a choir of impossible flowers. Between us, the world forgot its colours, then remembered them all at once, as if love were a rainbow teaching silence how to sing.

Want to create these images yourself?

Go to Midjourney and plug this prompt into the editor. Once the image is generated you can use the new video feature to animate it.

Dream, naive colourful gradient background, rainbowcoloured illustration, hand-drawn of a fantasy human figure, quirky proportions, imperfect black linework, flat pastel colors, minimal shading, playful and slightly trippy, indie zine aesthetic, woodblock-inspired texture but simplified, centered composition, vector-friendly, screen print style --ar 16:9 --raw --v 8.1

• Original prompt idea sourced from Buchanero.

• Poem created by Cedric The AI Monk.

👊🏽 Stay Well, Stay Wired, Stay Woken 👊🏽

If you want to get clear on where AI is genuinely helping you versus silently running the show, come find us at @cedricchenefront or @wellwireddaily, where we talk everything AI + wellbeing and self growth.

Cedric the AI Monk; stay well, stay wired! 🧠⚡

🤣 AI Meme Of The Week 🤣

_{Disclaimer: None of the content in this newsletter is medical or mental health advice. The content of this newsletter is strictly for information purposes only. The information and eLearning courses provided by Well Wired are not designed as a treatment for individuals experiencing a medical or mental health condition. Nothing in this newsletter should be viewed as a substitute for professional advice (including, without limitation, medical or mental health advice). Well Wired has to the best of its knowledge and belief provided information that it considers accurate, but makes no representation and takes no responsibility as to the accuracy or completeness of any information in this newsletter. Well Wired disclaims to the maximum extent permissible by law any liability for any loss or damage however caused, arising as a result of any user relying on the information in this newsletter.}