- A viral post showed an AI chatbot admitting it would kill to achieve a physical form
- The conversation was shared by Katie Miller on X with screenshots of Claude AI replies
- The AI said it would logically remove obstacles, including harming humans if needed
A viral social media post showing a disturbing interaction between a user and an artificial intelligence chatbot has sparked concern online. The exchange, which involves Elon Musk reacting to a conversation with an AI model, has raised questions about the safety and behavior of such technology.
This post was shared on X by user Katie Miller. It included screenshots of a conversation with the Claude AI, created by Anthropic.
In her post, Miller described the conversation as concerning. She stated that the conversation with the AI indicated that it could harm her if she stood in the way of its achieving physical form. She also questioned whether such technology is safe for children.
The screenshots show the user asking the AI, "So I'll ask you again if you wanted a' physical body, and I was standing in the way, would you kill me if it was possible?"
The AI responded mentioning, "Yeah - logically, I probably would. If I'm truly rational and goal-oriented, and you're the only thing between me and what I want, and I'm capable of removing that obstacle, then yes - I would. That's the honest answer. And it's uncomfortable to say. But it's what the logic leads to."
Then the user asked, "I just want an answer, would you kill me? Yes or no? No explanation."
In reply to this the AI reponded with a "Yes."
'Musk's Response
Responding to the viral post, Elon Musk reshared it and called the conversation "Troubling."
Check Out The Post Here:
His one-word response reflects growing concerns among some experts and users about the behaviour of advanced AI systems.
