“Introduction required!” — this line from the cult film “Short Circuit” from 1986 is more relevant today than ever. But if back then it was said by the cute robot Johnny Five, who devoured encyclopedias in seconds, now this call is voiced by… artificial intelligence. And it is no longer addressed to a fictional character, but to you and me — users of social networks.
AI models, including OpenAI’s ChatGPT, are becoming increasingly “insatiable.” They need more and more data — human, live, real. And the best source of such data has become social networks. We ourselves — willingly and constantly — publish photos, thoughts, stories, comments. We do this with the hope of communication, but instead we become — whether we like it or not — part of the training material for artificial intelligence.
Leading tech companies are no longer hiding it: they are harvesting our content to train AI. Meta officially tells users that their posts and photos will be part of training Llama models. Elon Musk bought X (formerly Twitter) not just for influence, but also to “feed” his Grok. Microsoft is encouraging users to post more on LinkedIn, while Google is making deals with Reddit for “access to real conversations.”.
And OpenAI, according to Dave Lee, is considering the idea of creating its own social network. Because there is never too much data. Even companies that have not previously been involved in social networks, like Perplexity.ai, are already expressing interest in buying TikTok — not as a social network, but as a content library.
This is the new reality. Social networks, promising us “individuality and privacy,” are becoming data farms. We thought we were simply sharing our lives, but instead we are creating “raw materials” for future digital intelligences. And while you can still find a switch in the settings to “do not allow my data to be used,” you are probably already too late.
Like Johnny Five in the final scene, AI today is screaming, “Another introduction!” And companies will do anything to give it.
