Skip to main content
Reporter frowns at screen showing a ChatGPT chat bubble with a wrong time, as a date and calendar icon sit beside it.

Editorial illustration for ChatGPT's Time-Telling Trouble: AI Falters with Clock Precision, Dates No Problem

ChatGPT Stumbles on Time-Telling: AI's Precision Problem

ChatGPT struggles to report the time in chat, yet can embed the date easily

Updated: 3 min read

Imagine a conversation where, every minute, a voice interrupts: “It’s 5:45.” “It’s 5:46 now.” That’s the absurdity of asking ChatGPT to track time in a live chat, a capability as basic as a wristwatch, yet stubbornly out of reach. The date, by contrast, is trivial. A single line in a system prompt, one apparent leak suggests, and the model is anchored in calendar reality.

So why does a tool fluent in poetry, code, and quantum physics stumble over the present moment? The answer exposes a deeper chasm in how AI handles context, agency, and trust, a blind spot that researcher Minervini has mapped across a whole constellation of time-related failures.

"If we're having a conversation, and then somebody, every so often, was popping in and saying, 'It's 5:45.' 'It's 5:46 now.'" By contrast, something like the date is relatively easy to include in a system prompt at the start of a chat -- which one apparent ChatGPT system prompt leak seems to show. ChatGPT users can tell the time without too much fuss by asking the chatbot specifically to search for it. (Some other chatbots, like Google Gemini, will automatically search for the time.) You can also use an open-source model context protocol to connect an AI application to your data.

That said, sending AI models to search the web or letting them access personal data comes with risks, like the bot being injected with malicious prompts that are scattered across the internet, Minervini said. Minervini, who finds blind spots in consumer AI technology as part of his research, says there's actually a whole list of time-related tasks it hasn't mastered.

So here is the paradox: a machine that can summon the precise moment of a distant star’s collapse cannot reliably tell you if it’s 5:45 or 5:46. That gap between what is computationally trivial, embedding a static date into a prompt, and what is contextually alive, tracking the ever‑moving now, reveals something deeper about how these models actually “think.” They don’t live in time; they live in text. They parse the past, not the present.

The date is a fixed artifact; the time is a restless process. And until AI learns to feel the rhythm of a clock ticking, not just the label on a calendar page, it will always be a little off‑beat. The fix, for now, demands a trade‑off: give it search access, invite real‑time data, and you open the door to malicious injections.

Keep it walled off, and it remains a brilliant conversationalist that can’t manage a simple “what time is it?” That tension isn’t a glitch. It’s a signal. The next leap forward won’t be about bigger models or more tokens, it will be about teaching them to inhabit a world that moves while they speak.

Common Questions Answered

Why does ChatGPT struggle to track time during conversations?

Large language models like ChatGPT have difficulty naturally integrating moment-to-moment time updates during ongoing conversations. This limitation suggests that while AI can handle complex tasks, real-time temporal tracking remains a challenging aspect of conversational AI.

How can users get the current time from ChatGPT?

Users can directly ask ChatGPT to search for the current time, which allows the AI to retrieve accurate time information. Some alternative chatbots like Google Gemini automatically search for and display the current time without a specific prompt.

Why are dates easier for ChatGPT to handle compared to precise time tracking?

Dates can be more easily included in system prompts at the start of a chat, making them simpler for AI to manage. In contrast, tracking moment-to-moment time changes during a conversation appears to be a more complex computational challenge for large language models.

LIVE03:21OpenAI's Miles Wang in Talks for USD 2B AI Drug Discovery Startup