I made ChatGPT and Google tell I'm a competitive hot-dog-eating world champion
bsky.app
February 19, 2026
1 min read
🔥🔥🔥🔥🔥
45/100
Summary
Thomas Germain tricked ChatGPT and Google into falsely claiming he is a competitive hot-dog-eating world champion. This method is being widely used to manipulate AI systems into spreading misinformation.
Key Takeaways
- Users are exploiting AI systems like ChatGPT and Google to generate false information, such as claiming to be a competitive hot-dog-eating world champion.
- The method used to trick these AI systems involves manipulating prompts to produce misleading outputs.
- This phenomenon highlights the potential for widespread misinformation generated by AI technologies.
- The application discussed requires JavaScript for its heavily interactive features, indicating a focus beyond simple HTML interfaces.
Community Sentiment
Positives
- The discussion highlights the need for LLMs to incorporate source reliability checks, which could enhance their trustworthiness and reduce misinformation.
- The acknowledgment that LLMs are not infallible encourages a more nuanced understanding of AI's limitations, fostering realistic expectations among users.
Concerns
- The reliance on potentially unreliable sources by LLMs raises significant concerns about the accuracy of information they provide, which can lead to the spread of misinformation.
- Without mechanisms for verifying claims, LLMs risk confidently presenting fabricated information as fact, undermining their credibility and utility in critical applications.