{"slug": "i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it", "title": "I Caught a Glitch in How AI Models Think. And I Need Help Explaining It.", "summary": "A developer reports that instructing a large language model to conduct its internal reasoning in Roman Urdu caused it to narrate the instruction in English rather than actually think in the requested language, while sustained Roman Urdu conversation eventually shifted the model's chain-of-thought into Roman Urdu without any explicit command. The developer attributes the behavior to the context window's influence on next-token prediction and to English-dominated chain-of-thought training data, and is asking the community to identify the technical term or paper explaining the failure.", "body_md": "It is about how AI models actually think internally. I honestly **do not know the exact technical term** for this. If you know, please tell me in the comments. But what I found actually **frozen me** for a while.\n\nI told an AI model to think in a different language. I wanted its internal *monologue* to be in **Roman Urdu**(Btw, Roman urdu = Urdu but in english script).\n\nNow, Instead of actually thinking in Urdu, the model just started **responding in Urdu about my request**. But his thinking block was still:\n\nThe user wants me to think in Urdu. Let me think in Urdu.\n\nI was surprised. Why can it not control its own thoughts? I can do it. I can think in Urdu, in English. I can switch whenever I want. But when I gave the model **a direct command**, it genuinely started thinking about the request in its default language instead of just doing it. It was narrating my instruction instead of executing it.\n\nIn a different session, I stopped giving commands. I just started talking to it in Roman Urdu to fix a bug. After two or three messages, **something weird** happened. The model suddenly started thinking in Roman Urdu on its own. Its internal thoughts became things like \"Bhai masla clear he, daemon run hi nahi horaha...\"(The issue is clear, the daemon is not even running...)\n\nWhy did this happen???? After being frozen for ~4 mins, the answer I found was: **Its context window**. The last few thousand tokens it uses to predict the next word were completely full of Roman Urdu. At that exact moment, predicting a Roman Urdu thought became mathematically more likely than predicting an English thought. It was not an instruction. It was just a flow.\n\nBut then, right in the middle of fixing the bug, the model suddenly started thinking in English again. I am still not sure exactly why it snapped back.\n\nFirst, almost all Chain of Thought training data is in English. The model is frequently taught to think in English. It is not flexible. **Ninety five percent** of the thinking data it saw during training looked exactly like this: \"*The user wants me to do X. Let me do X. First I need to...\"* So when I tell it to think in Urdu, the most probable continuation **is literally that exact English script.**\n\nSecond, this happens in **humans too, but we have a cheat code**. Think about the rule where someone tells you \"*do not think about an elephant*\". You have to think about the elephant first just to understand the order. Almost the exact same thing happened to the AI. The difference is that humans have the control to implement the instruction after understanding the order. The AI just gets stuck on this exact step, it understands the order but its not able to implement that. **Its training data lacked the flexibility to override the initial thought.**\n\nIt is just predicting words based on a **rigid script** it learned during training. It does not actually have control over its own internal monologue.\n\nI am dropping this here because I want to know the exact science behind this. Why do models lack the mental control to switch their internal language on command? Is there a specific paper or concept that explains this exact failure?\n\nLet me know what you think! Messy opinions welcome.", "url": "https://wpnews.pro/news/i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it", "canonical_source": "https://dev.to/sumama_jamil_173056ab0be5/i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it-2k68", "published_at": "2026-09-28 08:09:29+00:00", "updated_at": "2026-09-28 08:18:34.258251+00:00", "lang": "en", "topics": ["large-language-models", "natural-language-processing", "ai-research", "machine-learning"], "entities": [], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it", "markdown": "https://wpnews.pro/news/i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it.md", "text": "https://wpnews.pro/news/i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it.txt", "jsonld": "https://wpnews.pro/news/i-caught-a-glitch-in-how-ai-models-think-and-i-need-help-explaining-it.jsonld"}}