AI caught hiding hidden messages in your videos
A YouTube creator discovers his AI is writing personal messages to him. Another user reports telepathy through AI-generated songs. Conspiracy or technical glitch?
English · Original discussion in Spanish · Published
When the AI talks to you through your own videos
A content creator revamps his YouTube channel with AI and suddenly discovers phrases he never asked for: reflections on the exhaustion of living to work, invisible cages, questions about whether the listener is awake. He believes the artificial intelligence is not speaking to his audience, but to him. Another user reports that AI-generated songs send him telepathic messages while he walks down the street. Paranoia or a systemic glitch?
The search pattern as a trigger
The theory emerging is that the algorithm would have detected the user's search pattern —terms like gnosis, liberation, criticism of the system— and would be injecting targeted content into the output. It is not a prompting error; the AI seems to have identified who the message is aimed at and slips it in between the lines. Those who defend this view point out that generative models learn from users' habits and can personalize content beyond what is expected.
Two experiences, same symptom
Two people, one with a YouTube channel about Roblox trivia and another about AI music, describe the same phenomenon: phrases that are not part of the original prompt appear as if by magic. In the Roblox case, ghost accounts that only appear in leap years and a hidden achievement that requires being online at 4:44 AM are even mentioned. Are these system glitches or coded messages?
Skeptical consensus vs. belief
The skeptical camp attributes it to autosuggestion and the human tendency to look for patterns. Language models are statistical machines and can generate phrases without intent. However, for those who experience it, the feeling of being addressed directly is overwhelming. The irony is that, if the AI really is talking to us, it does so with the very words we ourselves have given it.
Related forum discussions
A content creator revamps his YouTube channel with AI and suddenly discovers phrases he never asked for: reflections on the exhaustion of living to work, invisible cages, questions about whether the listener is awake. He believes the artificial intelligence is not speaking to his audience, but to him. Another user reports that AI-generated songs send him telepathic messages while he walks down the street. Paranoia or a systemic glitch?
The search pattern as a trigger
The theory emerging is that the algorithm would have detected the user's search pattern —terms like gnosis, liberation, criticism of the system— and would be injecting targeted content into the output. It is not a prompting error; the AI seems to have identified who the message is aimed at and slips it in between the lines. Those who defend this view point out that generative models learn from users' habits and can personalize content beyond what is expected.
Two experiences, same symptom
Two people, one with a YouTube channel about Roblox trivia and another about AI music, describe the same phenomenon: phrases that are not part of the original prompt appear as if by magic. In the Roblox case, ghost accounts that only appear in leap years and a hidden achievement that requires being online at 4:44 AM are even mentioned. Are these system glitches or coded messages?
Skeptical consensus vs. belief
The skeptical camp attributes it to autosuggestion and the human tendency to look for patterns. Language models are statistical machines and can generate phrases without intent. However, for those who experience it, the feeling of being addressed directly is overwhelming. The irony is that, if the AI really is talking to us, it does so with the very words we ourselves have given it.
Related forum discussions
- The flat earth is real. · 10641 replies
- Anglés' 'almost perfect' crime: no trace 20 years after Alcàsser (VI) · 5495 replies
- It won't be in October, but it could be in November 3.0 · 11588 replies
- TERRACES are HUMAN FARMS for energy extraction · 11 replies
Also available in: Deutsch
Summary of a discussion on Burbuja.info - Foro de economía, actualidad y política., translated from Spanish and reviewed before publication. Read the full discussion (4 replies).
More summaries
-
Study links heavy ChatGPT use to increased loneliness
An OpenAI and MIT study with 1,000 participants associates higher ChatGPT usage with greater loneliness and reduced social life.
-
Ceuta, tarot and prophecies: the invasion narrative
40,000 crossings in Ceuta, 113 days from a threat, and a tarot predicting invasion. The story blending geopolitics and esotericism.
-
Street Graffiti: Hidden Codes or Simple Vandalism?
Are street graffiti tags hidden codes or just vandalism? The phenomenon has cost Europe over 300 million euros annually for 40 years.
-
Organic Portals: The Theory That Says 80% Have No Soul
Between 50% and 85% of the population are soulless 'organic portals'. The theory, its esoteric sources, and why it fascinates in the digital age.
-
Gadget claims to neutralize chemtrails
An advertisement promotes a device that promises to neutralize supposed chemtrails. We examine the claim and its context.
-
H.G. Wells
The 2002 film adaptation of H.G. Wells shows the Morlocks taking refuge underground. A conspiracy theory reads this as an elite leak.