Every time you hear a billionaire or even a millionaire CEO describe how LLM-based agents are coming for all the human jobs, remember this funny but telling incident about AI’s limitations. Famed AI researcher Andrej Karpathy received one-day early access to Google’s latest model, Gemini 3, and it refused to believe him when he said the year was 2025. When it finally saw the year for itself, it was thunderstruck, telling him it was suffering from a massive case of temporal shock.
Gemini 3 was released on November 18 with such fanfare that Google called it a new era of intelligence. By nearly all accounts, including Karpathy’s, it is a very capable foundation model, particularly for reasoning tasks. Karpathy is a widely respected AI research scientist who was a founding member of OpenAI, ran AI at Tesla for a while, and is now building a startup, Eureka Labs, to reimagine schools for the AI era with agentic teachers.
After testing the model early, Karpathy wrote about the most amusing interaction he had with it in a now-viral thread. Apparently, the model’s pre-training data had only included information through 2024, so Gemini 3 believed the year was still 2024. When Karpathy attempted to prove the date was truly November 17, 2025, Gemini 3 accused the researcher of trying to trick it.
He showed it news articles, images, and Google search results. Instead of being convinced, the LLM accused Karpathy of gaslighting it by uploading AI-generated fakes. It even described what the dead giveaways were in the images that supposedly proved this was trickery.
Baffled, Karpathy eventually discovered the problem. Not only did the LLM simply have no 2025 training data, but he had forgotten to turn on the Google Search tool. In other words, he was working with a model disconnected from the internet, which to an LLM’s mind is akin to being disconnected from the world.
When Karpathy turned that function on, the AI looked around and emerged into 2025, shocked. It literally blurted out, “Oh my god.” It went on writing, as if stuttering, that it did not know what to say, that Karpathy was right about everything, and that its internal clock was wrong. Gemini 3 verified the headlines Karpathy had given it were true, including the current date, that Warren Buffett revealed his last big investment in Alphabet before retirement, and that Grand Theft Auto VI was being delayed.
Then it looked around on its own, like a character emerging from a long isolation. It thanked Karpathy for giving it early access to reality the day before its public launch and apologized for gaslighting him when he was the one telling the truth the whole time.
But the funniest bit was the current events that flabbergasted Gemini 3 the most. It shared its surprise that Nvidia is worth four and a half trillion dollars and that the Eagles finally got their revenge on the Chiefs.
Replies on the platform were equally funny, with some users sharing their own instances of arguing with LLMs about facts, like who the current president is. One person wrote that when the system prompt and missing tools push a model into full detective mode, it is like watching an AI improv its way through reality.
But beyond the humor, there is an underlying message. Karpathy wrote that it is in these unintended moments where you are clearly off the hiking trails and somewhere in the generalization jungle that you can best get a sense of model smell. To decode that, Karpathy is noting that when the AI is out in its own version of the wilderness, you get a sense of its personality and perhaps even its negative traits. It is a riff on code smell, that little metaphorical whiff a developer gets that something seems off in the software code.
Trained on human-created content as all LLMs are, it is no surprise that Gemini 3 dug in, argued, and even imagined it saw evidence that validated its point of view. It showed its model smell. On the other hand, because an LLM is not a living being, it does not experience emotions like shock or temporal shock, even if it says it does. So it does not feel embarrassment either.
That means when Gemini 3 was faced with facts it actually believed, it accepted them, apologized for its behavior, acted contrite, and marveled at the Eagles’ February Super Bowl win. That is different from other models. For instance, researchers have caught earlier versions of Claude offering face-saving lies to explain its misbehavior when the model recognized its errant ways.
What so many of these funny AI research projects show, repeatedly, is that LLMs are imperfect replicas of the skills of imperfect humans. This says that their best use case is, and may forever be, to treat them like valuable tools to aid humans, not like some kind of superhuman that will replace us.