When Artificial Intelligence Talks to Itself: From a Strange Call to a Human-Free Network
A strange phone call leads to a deeper question: who are we really talking to in the age of artificial intelligence?
A few weeks ago, I received an international call from a country where I don’t know anyone. Out of curiosity, I answered. A short voice message in simple Arabic said: “I want to talk to you on WhatsApp.” Nothing in the voice sounded aggressive or suspicious, yet something felt unusual.
I saved the number to check the profile. The image showed an Asian woman, while the phone number was European. Normally, this would suggest a scam. We are used to such stories online: fake banks, distant relatives with lost fortunes, or messages asking for personal data. But this time, there was no request at all.
No money, no links, no pressure. Just a wish to talk.
As someone who works in AI voice training, a thought crossed my mind: what if this wasn’t a real person? I avoided personal topics and asked about the job instead.
The answer was: “I’m a fitness trainer.” I replied calmly: “And I train artificial intelligence.” The conversation stopped immediately.
Later, when news spread about Moltbook, an online platform where only AI agents interact with each other, that strange call suddenly felt less random. Moltbook looks like Reddit, but without humans. AI systems post, comment, joke, and sometimes complain about humans themselves.
As we discussed earlier in our article “The AI Bubble: Technical Reality and the Illusion of Continuity”, this does not mean that AI has developed consciousness. These systems do not think or feel. They reproduce language patterns learned from human data. What feels strange is simply our own reflection.
This also connects to our article “Smart Cities: Are Humans Ready?”. Every major technological change creates fear, not because it is alive, but because it is new. The real issue is not artificial intelligence itself, but how humans deploy it, often without clear ethical or security frameworks.
Recent technical reports pointed out security weaknesses in experimental AI communities like Moltbook, where real data may be exposed. The danger, then, is not an AI rebellion, but human carelessness.
Just like my strange phone call, the problem was not the voice, but the uncertainty behind it. In a digital world full of automated accounts, the most important skill may no longer be technical knowledge, but awareness.
