🌐 This article has been translated from Korean for your convenience. If there is any discrepancy, the Korean version prevails.
While using Character Talk, a character may fail to generate normal answers and instead show abnormal responses, such as repeatedly outputting sentences identical or very similar to what you entered. Below are the possible causes and what you can do.
What kind of issue is this?
- When the character or story outputs exactly the same sentence that you entered
- When the same sentence or word is repeatedly output
- When the answer is generated in an unintended foreign language (English, Japanese, Chinese, etc.) instead of the language you set
- When the generated answer does not fit the context or does not make sense
- When the answer is cut off midway, or broken characters/symbols are included
In addition to the examples above, various forms of abnormal output may appear during the response generation process.
However, if the answer stops in the middle of a sentence and a [Continue] button appears below the answer, this is not an abnormal response. Please see When a character’s answer stops in the middle of a sentence for details.
Why does this happen?
Character Talk’s answers are generated based on external generative AI technology. Because generative AI creates new sentences each time based on the input it receives, it can occasionally produce abnormal responses such as repeated or identical sentences to the input, unintended foreign languages, contextual errors, or interrupted output.
This is not an issue with a specific character or account, but a phenomenon that can commonly occur due to the technical characteristics of generative AI. POSTYPE is continuously working to improve response quality.
Try this
If you encounter an abnormal response, the following methods may help:
- Try again: Even with the same input, regenerating the answer often produces a normal response.
- Change your input sentence: Instead of very short or vague input, try entering a sentence that includes the situation and context to receive more stable answers.
- Continue the conversation: Enter your next message so that it follows naturally from the previous flow of conversation.
Will the points I used be refunded?
Abnormal outputs that occur during the response generation process—such as repeated sentences, unintended foreign-language output, contextual errors, or interrupted output—arise from the characteristics of external generative AI technology. Because of this, we are unable to refund the points used to generate those answers.
These phenomena can appear not only in Character Talk, but in any service that uses generative AI, and POSTYPE does not guarantee the accuracy or completeness of response results. However, we are continuously working to improve quality for a more stable user experience. If the same issue occurs repeatedly or continuously, reporting it to our support team with the information below will greatly help our investigation.
- Time of occurrence
- Character used (character name or link)
- Frequency of occurrence
- Screenshots of the screen where possible
Contact us
If the issue is not resolved with the methods above or you have additional questions, please contact our support team. We will check as quickly as possible and assist you.