Ai revolution: the shift from faster calculators to unified minds
The AI revolution is not just about having faster calculators, as Mustafa Suleyman, Microsoft's AI chief, warns. It's about achieving a unified, gigantic mind.
Exponential growth defies logic
Since 2010, the computing power used for training the most advanced AI models has grown by a trillion times. Suleyman likened the training process to a room full of people using calculators. In the past, AI meant packing more people into the room, resulting in many waiting idly for data. Now, the revolution lies in having thousands of calculators work in perfect sync as one giant mind.
This exponential growth, driven by three key pillars – ultra-powerful hardware, high-speed memory, and superconnection technologies – has led to a 700% increase in computing power per year. Microsoft and Meta are investing heavily in AI, but this rapid expansion comes with a critical challenge: energy consumption.
The future of ai agents
Currently, simple AI assistants are being left behind as we move into a new era with almost human-like AI agents. These systems won't just chat, but will work semi-autonomously for days or weeks, capable of managing complex projects, negotiating contracts, making calls, and organizing logistics for entire companies.
Microsoft and other tech giants are building massive data centers, investing billions of dollars to achieve the goal of Artificial Super Intelligence (ASI), a cognitive capacity that will far surpass human intelligence in almost every domain.
A critical obstacle: energy consumption
However, this massive deployment faces a critical hurdle: energy consumption. A single rack of AI servers consumes the same energy as 100 homes. Despite this, Suleyman remains optimistic about the decline in renewable energy costs, which have fallen almost 100 times over the past few decades. This opens the door for a future where it's easier to power these data centers.
As AI computing is expected to multiply by ten by 2027, it will be crucial to adapt and leave linear thinking behind. The AI revolution is underway, and it's up to us to navigate its complexities and shape its future.