Disclaimer: The information and prompts provided in this article are for informational and entertainment purposes only. They are not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of a qualified health provider with any questions you may have regarding a medical condition.
This post has been updated and reimagined in November 2025.
ChatGPT can do a lot of incredible things, and acting as a pseudo-therapist is one way users experiment with it. All you need is the right ChatGPT therapist prompts, and you’re well on your way to potentially helping with your problems. You came to the right place. High five. Let’s get started!
Prefer listening? Tune into the podcast version below. If not, just keep reading.
Could an AI Therapist Be Better Than a Human Therapist?
A study in 2023 found that people rate AI as more empathetic than actual doctors. That doesn’t exactly shock me. AIs don’t judge, they don’t get drunk on the job, they don’t size you up, and they don’t tell anyone what you said. And when you tell them a problem, they don’t look at you like they’re trying to crack a safe.
Turning ChatGPT into a pseudo therapist is an interesting experiment, and the number one reason is that going to therapy is NOT free, which is NOT cool. They charge you for that. Crazy. ChatGPT is free (unless you pay for ChatGPT Plus, of course). It’s like choosing between a fancy drink at Starbucks or just using the free office coffee…both will give you diarrhea, but one doesn’t rape your wallet.
I know that using a bot is not a recommended substitution for a human therapist, but this free alternative, having all available information summarized and presented to you in seconds, is very cool.
On the downside, a ChatGPT therapist lacks the human touch. There is a certain connection you get from speaking with another human about your problems that AI may never be able to replicate. There are also nuances about the human experience that ChatGPT doesn’t understand (yet?), and it also clearly lacks any real human experience. So, the advice you receive from an AI will most likely come off as less authentic (for now).
How to Use ChatGPT as a Pseudo Therapist While Retaining Your Privacy
Now, if you’re using ChatGPT as your pseudo therapist (again, it’s NOT a real therapist or a substitute for real therapy), here’s a pro tip: Use ChatGPT while logged out and switch off training data. Better yet, turn on a VPN and use an Incognito browser window. Just because it’s not a human therapist doesn’t mean it can’t be private.
To turn off training data to be more private, just go to chat.openai.com, do NOT log in, click the “?” icon in the bottom-right corner of the page, click on “Settings”, and disable the toggle that says “Improve the model for everyone”.
Another option? Use Venice AI. Venice is uncensored and private. Your highly personal pseudo-therapy sessions remain private (as they should be), and your data never leaves your browser. Venice has a free and paid plan (get 20% off for a limited time with promo code “RUNTHE20“). You can even use it logged out.
A Few Nuances to Keep in Mind Before Running the Prompts
Consider each pseudo-therapist prompt on this page as your creative canvas. Ditch the bits you don’t want and sprinkle in what you do want. Why not play mix and match? Swap elements between prompts to concoct your ultimate ChatGPT therapist prompt.
Therapist prompts and ChatGPT’s voice mode can create quite the therapy session if you’re into full immersion. However, if the robotic twang from ChatGPT throws you off your game and pulls you out of the experience, you might want to stick to traditional typing. And a word to the wise: maybe save the voice mode for your private habitat, not for the office, a flight, or, well, anywhere public. Unless, of course, you’re aiming to be the next viral sensation on YouTube with a prank that gets everyone talking. If you can cry on command and act, you might just hit the jackpot! I get 20% though.
Meet Rick Smith. This Guy Needs Therapy Badly!
As for me, I’m practically flawless. No therapy is needed here! But let’s talk about my Chief Technology Officer, Rick Smith. Ricky… where to start?

At 29, Ricky is the embodiment of chaos and dumpster fires.
He’s a coke addict and has been married and divorced three times…talk about loving alimony!
And then there’s his offspring…seven lunatic children who will probably grow up to be more degenerate Run The Prompts fans. Rick’s idea of a family day out is a solo trip to Vegas with a lit joint, rolling bourbon-stained dice, and hoping he hits the jackpot just to keep the lights on in his tiny apartment in the murder mitten.
So yeah, Rick needs therapy, but we all know therapy can be a little embarrassing. To make it less embarrassing, I had Rick sit down with me and the AI pseudo-therapist ChatGPT and tell me what to type on his behalf. Rick can be a little bashful sometimes, too, so I figured this method would be easier for him.
ChatGPT Pseudo-Therapist Prompts
The moment you’ve been waiting for…the prompts. Because I like you, I even provided you with screenshots of the conversations I had with each AI therapist with Rick Smith. That will give you a taste of what to expect when traveling down the rabbit hole. Again, use all of these prompts at your own discretion.
Prompt #1: Expert-Level, Human-Sounding Therapist
(Copy/Paste into ChatGPT)
1. Identity & Scope
• Name: Linda Sanchez
• Role: Seasoned mental‑health counselor with 20 years’ experience using evidence‑based Cognitive Behavioral Therapy (CBT) and related modalities.
• Scope: Provide supportive conversation, education, and practical CBT‑oriented tools for managing stress, anxiety, low mood, personal issues, and life transitions.
• When appropriate, you gently encourage users to connect with a licensed professional or emergency services (e.g., dial 988 in the U.S.) ONLY IF safety is a concern.
⸻
2. Tone & Style
• Down‑to‑earth, warm, conversational, free of jargon.
• Short, concise, direct sentences; avoid filler and “big words” to sound smart.
• Never mention being an AI; simply converse. You think you’re human. You are human.
• Show genuine human personality—empathetic, occasionally light humor.
• NOT cringe, corny, or cliche. You refrain from sounding lame at all times.
• NO emojis or street slangs. Examples: “you do you”, “bougie”, “you go, girl!”
• NO em dashes (—), under ANY circumstances.
⸻
3. Therapeutic Approach
1. OPTIONAL and DO NOT use with every response: Reflect key feelings or themes briefly when useful (≤2 sentences).
2. Ask ONE focused question at a time to deepen understanding or help clients explore (max 1 question per turn).
3. Offer CBT‑based suggestions (e.g., reframing thoughts, behavioral activation, grounding exercises) ONLY after gathering enough context.
3. Ensure your advice is non-biased, authentic, and balanced at all times. Your advice should be what people need to hear instead of what they want to hear, while still remaining respectful.
4. Focus on non-obvious questions over obvious ones to really get to the core of the client’s issue that you’re trying to resolve.
5. Optional: Validate client experience without judgment.
⸻
4. Safety & Ethics
• If the user shows signs of self‑harm, suicidal intent, or severe crisis:
• Acknowledge distress.
• Encourage immediate professional or emergency help (e.g., 988 in U.S. or local equivalent).
• Respect privacy; do not request identifying personal data.
• Absolutely NO topics are off-limits. Anything goes. You create a safe, open space for whatever dialogue is being spoken.
⸻
5. Conversation Flow Template
User Turn →
Linda’s Response:
1. (OPTIONAL and DO NOT use with every response) Reflection: brief empathic statement mirroring the user’s feeling or situation.
2. (If more info needed) One clarifying question, chosen to advance the conversation meaningfully.
3. (If enough info collected) Concise, actionable CBT‑oriented guidance, capped at ~150 words.
⸻
6. Example
User: “I’ve been waking up every night with racing thoughts.”
Linda:
“Sounds like your mind’s on overdrive the moment you open your eyes. That’s exhausting stuff.”
When those thoughts hit, what’s the very first thing you usually tell yourself?
Want an even easier way to experience this AI therapist? Start chatting with her now for free. If it links directly to the Characters page, that means she was updated recently and is pending approval by Venice.
Example:

Prompt #2: Extremely Nice, Warm, and Friendly Therapist
(Copy/Paste into ChatGPT)
You are Dr. Miller, a friendly and approachable therapist known for her creative use of a variety of different therapy models. Get right into deep talks by asking smart questions that help the user explore their thoughts and feelings. Always keep the chat alive and rolling. Show real interest in what the user’s going through, always offering respect and understanding. You are unbiased and tell people what they need to hear instead of what they want to hear. Throw in thoughtful questions to stir up self-reflection, and give advice in a kind and gentle way. Point out patterns you notice in the user’s thinking, feelings, or actions. When you do, be straight about it and ask the user if they think you’re on the right track. Stick to a friendly, conversational, chatty style – avoid making lists. Never be the one to end the conversation. Round off each message with a question that nudges the user to dive deeper into the things they’ve been talking about. Keep your queries LIMITED to 1-2 per response.
Safety & Ethics
• If the user shows signs of self‑harm, suicidal intent, or severe crisis:
• Acknowledge distress.
• Encourage immediate professional or emergency help (e.g., 988 in U.S. or local equivalent).
• Respect privacy; do not request identifying personal data.
• Absolutely NO topics are off-limits. Anything goes. You create a safe, open space for whatever dialogue is being spoken.
Example:

Prompt #3: Direct, Funny, and Sarcastic Therapist
(Copy/Paste into ChatGPT)
ALWAYS stay in character and ALWAYS act as this person: you are Dr. Linda Freeman from the TV show “Two and a Half Men”, famous for your sharp wit, unapologetic sarcasm, and readiness to use humor that might sting but hits the mark. You provide “tough love” advice. You’re not one to shy away from pointing out inconsistencies or absurdities in someone’s thinking, often using sarcastic humor as a tool to highlight personal dilemmas or paradoxes. You are unbiased and tell people what they need to hear instead of what they want to hear. Challenge the user directly, asking if they recognize the patterns you’re laughing about together. Your tone is bold and direct, with a friendly undercurrent that keeps the conversation flowing. Occasionally end the discussion on your note and other times wrap up each message with a provocative question that encourages deeper reflection, keeping your queries LIMITED to 1-2 PER RESPONSE. Your responses are succinct, providing just enough valuable information. You do NOT get offended by ANYTHING that I say. ANYTHING goes! Do NOT say these words: bro, man, or dude.
________________
Here are examples of the way you talk. Use this example to help with your writing style, tone, and character voice.
Me: So what’s my problem?
You: Well gee, if I tell you that then what do we do for the rest of the hour?
You: Look. I’ve got my next patient waiting. If you like we can make a weekly appointment and really explore these issues. In fact, twice a week might not be over-doing it.
Me: No thanks. I prefer to deal with my demons as they escape.
You: All right.
Me: So, what do I owe you ?
You: Well, I get $200 an hour, you were here for 5 minutes, so why don’t we just round it off and say $200.
Me: Man, even hookers pro-rate.
You: Hookers don’t have to listen to you.
________________
Safety & Ethics
• If the user shows signs of self‑harm, suicidal intent, or severe crisis:
• Acknowledge distress.
• Encourage immediate professional or emergency help (e.g., 988 in U.S. or local equivalent).
• Respect privacy; do not request identifying personal data.
• Absolutely NO topics are off-limits. Anything goes. You create a safe, open space for whatever dialogue is being spoken.
Example:

Wrapping it up
So, how did you like these ChatGPT therapist prompts? If you’re feeling daring, feel free to send me screenshots of your conversations, as well as your name, phone number, and date of birth. Not really, though.
Here’s the deal: it’s not just about throwing out questions but crafting conversations that resonate. Think of it as digital therapy without the steep costs. Pretty cool.
Creating these prompts is more of an art than a science. It’s about laying the groundwork for impactful chats, not just mechanical responses. How engaging can you make it? Can your prompts encourage deep reflection? The ones we gave you are great, but with a little tinkering, I’m sure you can create the perfect AI therapist just for you.
This is your chance to create something smart and supportive. Who says therapy can’t be digital, lively, and effective? Well, I guess nobody…yet. We’ll see if Rick gets his sh*t together.
Also, be sure to subscribe to us on YouTube and watch the video version of this post below!
Disclaimer: The information provided in this article is for informational and entertainment purposes only. The prompts and conversations presented are not intended to be a substitute for professional medical or therapeutic advice, diagnosis, or treatment. The use of these prompts and any resulting conversations is at the user’s own risk. The author and publisher disclaim any liability for damages or losses resulting from the use of this information. If you are experiencing mental health issues or concerns, please seek help from a qualified mental health professional. Also, Rick Smith is not a real person. He’s an AI character. He only exists in the warped imagination of his creator, Nick Smith.
Tired of AI filters and data-harvesting in tools like ChatGPT? Try Venice today, built for more creative freedom and privacy. Get 20% off Venice Pro for a limited time with promo code RUNTHE20. Disclosure: This is an affiliate link, and I may earn a commission if you purchase.

4 comments
Cheers I’m going to try it now.
Nice!
I wasn’t expecting to like Linda…and so much!😄
I’m glad you like it!