Users in the US have expressed surprise and concern after ChatGPT has begun occasionally including words in foreign languages – including Arabic – in its responses.
Over the past month, many users have posted examples on social media where the popular artificial intelligence chatbot provides answers in English, but suddenly inserts words or parts of words in Arabic, as well as other languages such as Armenian, Hebrew, Spanish, Chinese, and Russian.
Some users have also reported that even the numbers in the response have been displayed in different language formats, making the text confusing and unexpected.
A user on Reddit wrote that the system had given him recipe ingredients where some words had appeared in Arabic, for no apparent reason and unrelated to the language of the request.
Possible explanation: How the model works
Experts say the phenomenon is not related to "conscious errors" of the system, but to the way artificial intelligence models work.
ChatGPT is a large language model that doesn't read sentences like humans do, but rather breaks them down into small units called "tokens." These can be whole words, parts of words, or even signs from different languages.
In some cases, the system may choose a shorter or more efficient “token” to construct the answer, even if it belongs to a different language. This does not mean that the AI is “intentionally switching languages,” but that it is choosing the most statistically likely variant.
What is ChatGPT and how is it used?
ChatGPT, developed by the company OpenAI, is one of the most widely used artificial intelligence systems in the world, with hundreds of millions of monthly users.
It is used for writing texts, translations, explanations, content creation and problem solving. Its competitors include systems like Google Gemini, Grok and Claude, but ChatGPT remains among the most dominant in the market.
Is it a new problem?
According to reports, similar problems have been observed before, including cases where the system produced incomprehensible text after technical updates.
However, OpenAI has not yet provided a specific official explanation for the recent cases of language mixing.
User feedback
Some users have described the phenomenon as disturbing and inexplicable, emphasizing that they had not encountered such cases before.
One user wrote on social media that he was "confused" why the system was responding to him in a different language, considering this a strange and unexpected error.
In general, experts emphasize that this is a technical effect of the way AI models process language, and not a deliberate change in the system's behavior. /GazetaExpress/
t7 live
