Former Google CEO Eric Schmidt shared candid insights in an exclusive TED interview about what constrains artificial intelligence development and what challenges await humanity in the near future.
G. Ostrov
Eric Schmidt, former CEO of Google, presented a comprehensive analysis of artificial intelligence's current state and future prospects in a recent TED platform interview. According to him, society still fails to fully grasp the power and potential of AI technologies.
AlphaGo's Historic Victory as a Turning Point
The artificial intelligence revolution began in 2016 when DeepMind's AlphaGo system accomplished something incredible: it made a move in the game of Go that no human had conceived of in 2,500 years of the ancient game's existence. This moment became the first convincing proof that AI could not only mimic human thinking but also create fundamentally new solutions.
ChatGPT — Just the Tip of the Iceberg
Schmidt emphasizes that the popular ChatGPT represents only a small fraction of modern AI capabilities. Real breakthroughs are happening in reinforcement learning, strategic planning, and multi-step problem solving. Contemporary systems can already independently perform complex operations requiring long-term planning.
Limitations and Future Goals
Despite impressive achievements, AI currently struggles with interdisciplinary research and discoveries that come easily to human scientists. The next ambitious goal is to teach artificial intelligence to independently formulate objectives and create new scientific paradigms.
Three Key Problems in AI Development
According to Schmidt, artificial intelligence growth is constrained by three critical factors:
Energy Crisis
The United States alone will soon need an additional 90 gigawatts of electrical power — equivalent to building 90 nuclear power plants exclusively to power AI data centers.
Data Shortage
Humanity is approaching the depletion of quality data for training AI systems. The solution may be creating synthetic data, but this requires fundamentally new approaches.
Computational Power
Transitioning to full reinforcement learning will require a thousand times more computational resources compared to traditional deep learning.
Geopolitical Risks and the Race for Supremacy
Schmidt draws attention to the dual nature of AI technologies, which can be used for both peaceful and military purposes. Today, the US and China are competing for dominance in the artificial intelligence sphere. The first country to achieve superintelligence level will gain technological superiority capable of destabilizing the global balance of power.
The former Google CEO compares the modern AI race to the nuclear deterrence period of the Cold War, but with one critical difference — the margin for error is significantly smaller.
Call to Action
"If you're not using artificial intelligence — you will inevitably fall behind. Learn it every day. This isn't a temporary trend, but a lifelong marathon," Eric Schmidt urges everyone who wants to remain competitive in the future.
Learn more about artificial intelligence and its impact on society at the official TED website.
If you have any problems, contact us — we'll help quickly and efficiently!