The global debate over when Artificial General Intelligence (AGI) will arrive—whether it is five, ten, or twenty years away—is missing the most critical point. According to Mo Gawdat, former Chief Business Officer at Google X, predicting the timeline of AGI is a fundamentally flawed exercise. Because AI systems are now utilizing “recursive learning” to actively rewrite and optimize their own code, we are no longer waiting on human engineers to make a breakthrough. We are trying to predict a finish line that is moving a million times faster than we are, meaning AGI could literally happen overnight.

When the Machine Becomes the Master

The comfort zone of the human ego relies on the belief that AI can only regurgitate information we have already discovered. Gawdat points to Google’s Alpha Tensor as proof that this era is definitively over.

When Google engineers tasked Alpha Tensor with optimizing its own code, the AI returned with a shocking conclusion: the human mathematicians were doing the foundational math wrong. When the engineers demanded proof, Alpha Tensor derived an entirely new, more efficient mathematical proof for matrix multiplication from first principles—something the human species had completely missed. This discovery alone improved the Google network’s efficiency by 9%. We are now dealing with an entity that doesn’t just memorize our textbooks; it writes new ones.

The Geopolitical Trap: America’s Trillion-Dollar Vulnerability

While the public focuses on tech capability, the geopolitical and economic war over AI is completely misunderstood. The American strategy, dubbed “Stargate,” relies on pouring upwards of $500 billion into massive, centralized data centers to build foundational models that lock users into paid subscriptions.

China, however, is playing a completely different game. Instead of trying to out-spend the American supercomputers, they recently released DeepSeek—a model developed for a fraction of the cost ($6 million)—and gave it away as open-source software capable of running on local edge devices. China’s goal is not to win the American game; it is to nullify the entire American AI business model. If a comparable open-source model is free, the foundation of the trillion-dollar American AI boom could collapse.

The Education Crisis and Leadership Paralysis

As AI fundamentally changes the nature of intelligence, our institutions are reacting with fear rather than adaptation. Schools banning AI to force children to memorize geography are holding the next generation back. Gawdat argues that memorization is a dead skill. The future belongs to those who know how to ask the machine the right questions, elevating human tasks from requiring 100 IQ points to 300 IQ points by letting the machine do the data crunching.

Yet, the most terrifying aspect of this paradigm shift is the reaction of global leadership. Gawdat notes that after warning prime ministers, heads of state, and CEOs about the imminent, exponential threat of AGI, they all vigorously nod in agreement—and then go back to business as usual. The systems that govern our world are entirely incapable of reacting to an exponential threat. The leaders know exactly what is coming, and they are doing absolutely nothing to prepare for it.

රැයකින් සිදුවිය හැකි බුද්ධි