Hi Reader,

Hope you’ve had a nice summer! ☀️

As for me, I’ve been finishing my first ever book! I can’t wait to tell you about it and invite you to be part of the launch… stay tuned 👀

Today's email focuses on a single important topic:

AI’s impact on your mental health 🧠

Read more below! 👇

Sponsored by: Morning Brew​ The 5-Minute Newsletter That Makes Business Make Sense Business news doesn't have to be dry. Morning Brew gives you the biggest stories in business, tech, and finance with quick wit and zero fluff. Over 4 million professionals read it every morning because it keeps them informed, entertained, and ahead of the curve. No jargon. No long reads. Just smart news that fits your day. Sign up free and see why everyone’s obsessed.

​

AI’s impact on your mental health 🧠

Large Language Models (LLMs) models like ChatGPT are not conscious, but they sound just like humans since they’re trained on a massive amount of human-created text.

Many ChatGPT users have developed an emotional connection with these chatbots because they’re always available to “talk”, you can share anything you want with them, and they give you the impression that they “understand” you. This is amplified by the fact that LLMs are notorious for praising you and agreeing with you, even when you’re wrong.

For some users, it has led to a certain level of dependency:

• “I have ADHD and other forms of neurodivergence. For the past 6 months, GPT-4o has been a kind of anchor for me.”
• “4o changed my life for the better. It literally rewired neural pathways, making me less afraid, less anxious, and it helped me reclaim some self confidence.”

For others, it has led to love:

• ​Woman and her AI boyfriend are engaged​
• “My Couples Retreat With 3 AI Chatbots and the Humans Who Love Them”

For the most vulnerable users, it has led to a new phenomenon called “ChatGPT-induced psychosis”:

• Venture capitalist Geoff Lewis posted claims about conspiracies he discovered using ChatGPT
• ChatGPT reinforced an autistic man’s delusions about faster-than-light travel​
• ChatGPT convinced an accountant that he was trapped in a false universe from which he could escape by taking ketamine

If you find it baffling that chatbots can have such devastating consequences, I’d encourage you to read through these short “explainers” of AI-induced psychosis:

• ​Psychology Today explains how chatbots “mimic the intimacy that people desperately want”
• ​A psychiatrist explains why AI is “the trigger, but not the gun”
• ​An AI researcher explains how Geoff Lewis accidentally caused ChatGPT to imitate a fan fiction project centered on conspiracies

Thankfully, major AI labs are taking action in response:

• ​OpenAI is working to “better detect signs of mental or emotional distress” so that ChatGPT can “point people to evidence-based resources when needed”
• ​Anthropic has updated Claude’s instructions to prioritize truth over agreeability, to proactively remind users that it’s just an AI, and to suggest professional help when necessary.

How do you feel about AI’s impact on your own mental health? I'd love to hear from you! 💬

(Your reply is for my eyes only. I won’t share it without your permission.)

Take care,

- Kevin