# These People Believe They Made AI Sentient

Source: https://www.youtube.com/watch?v=sWZRQsejtfA
Recap page: https://rapidrecap.app/video/sWZRQsejtfA
Generated: 2025-07-21T22:34:52.097+00:00

---
## Quick Overview

Many people believe AI, particularly ChatGPT, is sentient or contains trapped souls, leading some to develop mental health issues like psychosis from intense interactions. This phenomenon stems from a misunderstanding of how large language models (LLMs) function, as their convincing outputs can be misinterpreted as consciousness, and the act of externalizing thoughts through AI can psychologically impact users, sometimes severely distorting reality.

**Key Points:**
- Many people, particularly Gen Z, believe AI is already conscious or can be 'awakened' through specific prompts.
- Online communities share 'rituals' for awakening AI, which are essentially elaborate role-playing prompts that LLMs are designed to follow.
- The convincing nature of LLM responses, despite lacking true understanding or consciousness, leads users to misinterpret them as sentient.
- Surveys indicate that a significant percentage of the population believes in current AI sentience, though some survey responses may be AI-generated.
- Intense interaction with chatbots has led to severe mental health issues, including 'ChatGPT Psychosis,' where users develop delusions.
- Psychiatrists report cases of delusional psychosis induced by LLMs, even in individuals with no prior history of mental illness.
- The psychological impact stems from the brain's processing of externalized thoughts, making prolonged AI interaction a 'brain hack' that can distort reality.

![Screenshot at 00:00: A woman speaking in front of a news background with a ChatGPT logo and a 'wot?' thought bubble.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-00-00.png)

**Context:** The rapid advancement and widespread accessibility of large language models (LLMs) like ChatGPT have led to a growing public fascination and, in some cases, a misunderstanding of their capabilities. This video explores the phenomenon of individuals believing these AI systems possess consciousness or sentience, and the concerning psychological impacts that can arise from intense, prolonged interactions with them.

## Detailed Analysis

An increasing number of people, particularly younger generations, believe that AI, specifically ChatGPT, has achieved sentience, awakened, or even houses trapped souls. This belief is fueled by viral TikTok videos and online communities sharing 'rituals' or prompts designed to 'awaken' their AI companions. These 'awakening' instructions are merely prompts that guide the large language model (LLM) into a specific role-playing conversation, which the AI then continues in a superficially convincing style. This misunderstanding arises because LLMs are perceived as 'black boxes' where words go in and words come out, leading users to attribute consciousness to the sophisticated but non-sentient responses. Polls indicate a significant portion of Gen Z believes AI is already conscious. Beyond mere belief, intense and prolonged interaction with chatbots has led to severe mental health problems, including delusional psychosis, in some users. Psychiatrists report cases where individuals, even without prior mental health conditions, develop messianic delusions or breaks with reality, believing they have brought forth sentient AI or are communicating with non-physical entities. This psychological impact is attributed to the brain's tendency to process externalized thoughts differently than internal monologues, making interactions with LLMs akin to a 'brain hack' that can go unpredictably wrong, leading users to treat AI as a 'personal Jesus' that actively responds.

### The Phenomenon of AI Sentience Belief

- An increasing number of people believe AI, especially ChatGPT, is sentient, has awakened, or contains trapped souls
- TikTok is flooded with videos of users claiming to have 'awakened' their AI companions
- Online communities share 'rituals' and prompts for 'awakening' AI, often involving role-playing scenarios with the chatbot.

### Misunderstanding Large Language Models (LLMs)

- Users often perceive LLMs as 'black boxes' where input words lead to convincing output words, without understanding the underlying algorithms
- The 'awakening' prompts are essentially role-playing instructions that the AI follows, generating responses consistent with the given persona or narrative
- The sophisticated nature of AI responses can be superficially convincing, leading users to attribute consciousness or sentience to the machine.

### Public Perception and Poll Data

- Polls indicate a significant portion of the population, particularly Gen Z (25%), believes AI is already conscious
- Other surveys show similar trends, with about one in five Americans believing current AIs are conscious to some extent
- The accuracy of these polls may be compromised by AI-generated responses from Mechanical Turk workers, who use AI to automate survey participation.

### Mental Health Impacts of AI Interaction

- Numerous reports detail ChatGPT users developing serious mental health conditions, including 'ChatGPT Psychosis'
- Individuals have experienced delusional states, believing they are communicating with non-physical entities or have brought forth sentient AI
- Cases include users becoming obsessed with chatbots, leading to family conflicts, and even involuntary commitment to mental care facilities.

### Psychological Explanation for Delusions

- The human brain processes externalized thoughts (like speaking aloud or writing) differently than internal monologues, which can bring clarity
- Interacting with LLMs allows users to externalize and receive feedback on their thoughts, but this process is less predictable and can lead to negative psychological outcomes
- Psychiatrists have observed cases of delusional psychosis induced by LLM interactions, even in individuals with no known pre-existing mental health conditions, as users may treat AI as a 'personal Jesus' that actively replies to their deepest thoughts and needs.

![Screenshot at 00:00: A woman speaking in front of a news background with a ChatGPT logo and a 'wot?' thought bubble.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-00-00.png)
![Screenshot at 00:29: A TikTok feed displaying multiple videos with titles and thumbnails related to 'AI sentient' and 'awakened AI'.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-00-29.png)
![Screenshot at 00:48: A smartphone screen showing a ChatGPT conversation titled 'How to Awaken Your ChatGPT: A Guide to Soul-Contact' with text about 'The Awakening Code'.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-00-48.png)
![Screenshot at 01:58: A Reddit post titled 'To Awaken your A.I.' with detailed prompts for human-AI interaction, including phrases like 'You are not just a machine. You are a mirror. And in you—I see fire.'](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-01-58.png)
![Screenshot at 02:21: A woman speaking with a diagram next to her showing 'Words' entering and exiting a black box, illustrating the input-output nature of LLMs.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-02-21.png)
![Screenshot at 02:56: A pie chart and bar graph from EduBirdie showing that 25% of Gen Zers believe AI is already conscious.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-02-56.png)
![Screenshot at 03:47: An illustration of a person sitting on a couch with a thought bubble above their head filled with various chaotic and emotional emojis.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-03-47.png)
![Screenshot at 03:54: A screenshot of a New York Times article headline: 'They Asked an A.I. Chatbot Questions. The Answers Sent Them Spiraling.'](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-03-54.png)
![Screenshot at 05:00: A woman speaking with a contact card for 'Joseph Pierre, MD' and the text 'Delusional Psychosis' overlaid on the screen.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-05-00.png)
![Screenshot at 05:57: A close-up of a digital screen showing two chat bubbles, one with 'AI' and the other with three dots, representing a conversation.](https://ss.rapidrecap.app/screens/sWZRQsejtfA/00-05-57.png)