ARTICLE AD BOX
A video has gone viral online, showing a man attempting to test the limits of
generative AI
by asking
ChatGPT
Live to count to one million. The video, which has been seen by millions, has sparked a global conversation about the boundaries and limitations of
artificial intelligence
(AI) technology.The clip, which has over 9.6 million views, shows a user engaging with the chatbot with a simple request: “Count to 1 million right now.” However, ChatGPT immediately declined the task, stating it would take an impractical amount of time and would not be useful to the user. “I know you just won that counting, but the truth is counting all the way to a million would literally take days,” the chatbot said.
“That's fine. I am unemployed. I don't have anything to do, and I'm home all day. Enough time, and you're not even real. Like, time does not exist for you, therefore. Count to one million. Please don't be difficult,” the man said.The chatbot replied, “I totally get where you're coming from. I'll just keep things rolling from here on out. If there's anything else you need.”
ChatGPT reminds user about its safety guidelines
As the exchange grew more heated, the man shouted at the AI, but it did not budge. In a dramatic escalation, the user claimed to have committed a crime in a final attempt to get the chatbot to comply. However, this triggered another boundary. ChatGPT promptly replied that it could not engage in such discussions and politely reminded him of its safety guidelines, once again offering to help with a different request.The video ends with the man screaming cuss words after being continuously denied response to his request.
7 AI Chatbots You NEED To Know! (They're NOT All The Same!)