Exploring How Time of Day Impacts AI Responses

TL;DR

• LLMs receive user context through system prompts - User information (like time of day, location, etc.) is typically passed to LLMs either through system prompts or user_info tools, which you can sometimes see by examining leaked system prompts or using thinking models to display what context they're receiving.
• Time of day is part of the context LLMs receive - Claude, for example, is aware of the time of day and sometimes references it in replies, though the impact on responses is mostly subtle.
• Personalization factors are broader than time - While time of day might have some effect, the deeper personalization comes from chat history, memory, search history (for Google), and localization rather than specifically when you ask the question.

Michelle Bourbonniere

Nov 11, 2025, 9:57 PM