Sam Altman decided to respond to the criticism of the ecological footprint of artificial intelligence provocatively. At an event in India, he declared that comparing the power consumption of an AI to a human is “unfair” because even training a human costs a huge amount of resources.
“It takes about 20 years of life and all the food you eat during that time to become wise,” he said.
According to him, evolution and the billions of ancestors who created civilization must also be taken into account. writes Futurism. From this point of view, it is said that “AI has probably already caught up with humans in terms of energy efficiency”.
So far, nothing even comes close to the energy efficiency of the brain
However, such an argument runs counter to the reality of 2026. The first weakness is the technology itself. Cutting-edge research today focuses on neuromorphic chips, processors that mimic the functioning of the human brain, because it is still true that the brain is extremely frugal.
The human brain operates on approximately 20 watts. That’s less than a light bulb in a refrigerator. Training GPT-5 type models or their successors consumes millions of kilowatt hours. Technology companies invest in nuclear resources and huge data centers precisely because current AI is energy intensive.
If AI were truly more efficient than humans, it would not need its own power plants. The very fact that developers are exploring the brain as inspiration to reduce consumption, it represents silent evidence that the biological system is winning for now.
Don’t overlook
The popular news app for iOS and Android has received a major update. A lot of new features have been added that you must try
The second problem lies in the so-called “evolutionary debt” of artificial intelligence. Altman counted 100 billion ancestors in the energy balance of man. But AI does not exist in a vacuum. The models were created on the basis of a digital record of human civilization, books, code, scientific articles, art.
This data was created over thousands of years and digitization took decades. Servers, fiber optic cables, cloud infrastructure all cost energy and resources. So if we count the evolution of man, we must also calculate the energy footprint of the entire digital infrastructure, which enabled model training.
Without this “human gift”, AI would be just an empty algorithm. The model did not learn the language by itself. He learned it from the work of generations of authors, programmers and scientists.
Altman also responded to the question of water. He called the claims about tens of liters per question absurd.
“It’s completely false and totally insane. It has no connection to reality,” he declared.
However, the debate about the exact numbers continues, because tech companies only release limited usage data.
Who of whom
Behind these statements there is also a psychological shift. Just a few years ago, AI was described as a “co-pilot” or “assistant”. The language emphasized cooperation. However, Altman’s statement in February put humans and AI in a competition for resources.
When someone says that a person takes a long time to study and “eat a lot of food”, it gives the impression that human biology is a cost problem. in the environment, where there is a fight for electricity for data centers and households face rising energy prices, such an assertion falls on sensitive ground.
So the question is not just how many watt-hours it costs to train the model. The question is what the company considers to be a priority. Should energy go to hospitals, homes and infrastructure, or to ever larger models?
Altman tried to move the discussion from data centers to the plate of food. But in doing so, he opened a wider debate about the value of a person in an era where technology is becoming a competitor, not just a tool.
AI can bring progress, optimization but also new discoveries. However, the human brain remains the most efficient computing system we know. The dispute about the energy balance will not end with one sentence from the podium, but with specific numbers and decisions about where the company directs its resources.
