Sam Altman, Chief Executive Officer of OpenAI, has once again shaken the world with his statements. This time, it was not about prospects or theoretical models. Altman stated outright what many have been afraid to say aloud: “We may have already crossed the threshold where artificial intelligence became smarter than humans.”
And yes, this is not a hypothesis. According to the OpenAI head, the intellectual capabilities of some AI systems have already exceeded the average levels of human intelligence in certain areas. However, this is not yet apparent to everyone — AI still appears externally as a familiar assistant, a text or image generator. But under the hood, it possesses power that, according to Altman, can rival the combined cognitive capabilities of a vast number of people.
To put it bluntly: we are living in an era when artificial intelligence may already be smarter than most of us — and continues to grow rapidly.
Altman predicts that by 2026, AI systems capable of making genuine scientific discoveries will emerge, and by 2027, robots with physical embodiment will be able not only to follow instructions, but to act autonomously in the real world. According to him, by 2030, the world will experience a radical leap in the accessibility of intelligence and energy — something he calls the “democratization of computational power.”
What does this mean? That ideas which today require billions in funding and hundreds of engineers may, within five years, be executed by a team of five — with the support of AI.
Naturally, such statements have been swiftly picked up by skeptics. Some recall elections, manipulation, deepfakes, and cybersecurity threats. Others argue that this is part of OpenAI’s marketing strategy ahead of its next funding round. But something suggests: this time, Altman is not merely provoking.
After all, if one thinks about it… Perhaps we really did not notice when the machines became smarter? While we are arguing — they are already writing code, inventing medicines, designing buildings, forecasting markets, and modeling climate.
And so, the main question is not even “when.” The main question is — who is in control here?
