That question came up in The Minds of Modern AI, a roundtable with Geoffrey Hinton, Yoshua Bengio, Yann LeCun, Fei-Fei Li, Bill Dally, and Jensen Huang — six pioneers who built the foundations of modern AI.
Their responses were thoughtful, but notably optimistic.
🔹 Jensen Huang compared today’s AI boom with the dot-com era. Then, much of the fiber network was unused, the so-called “dark fiber.”
Today, he said, “almost every GPU you can find is lit up and used.” AI isn’t speculation and it’s production. GPUs are now “factories of intelligence,” powering industries worth trillions. “We’re at the beginning of the build-out of intelligence,” he said.
🔹 Bill Dally argued that we’re still early in the curve. Models are getting more efficient, more capable, and more widely applied, maybe only 1% of potential demand has been reached.
🔹 Fei-Fei Li reminded that AI is a young science, only 70 years old compared to 400 years of physics. She sees vast unexplored frontiers such as spatial intelligence, which connects perception and action, areas where today’s language models still fail.
🔹 Yann LeCun said the real risk isn’t financial but conceptual. Scaling current LLMs won’t lead directly to human-level intelligence: “We don’t even have robots as smart as a cat.” Real breakthroughs will need new science.
🔹 Yoshua Bengio agreed there could be short-term corrections, but long-term, the growth is justified. He called for international collaboration to ensure AI evolves safely and ethically.
🔹 Geoffrey Hinton echoed that tone of caution. He warned that even if AI’s progress is real, our control over it may not be. He urged experts to continuously track AI’s trajectory, risks, and mitigation efforts.
Listening to them, I couldn’t help but think, perhaps none of them would publicly say we’re in an AI bubble, especially while receiving one of the highest honors in engineering.
But that doesn’t mean a bubble isn’t forming.
From a broader view, valuations, media attention, and corporate spending all seem inflated relative to adoption and understanding.
Yet their arguments remain valuable: even if we are in a short-term AI bubble, their reasoning highlights why AI will still define the next decades once the hype settles.
So maybe the truth is both.
Yes, we might be in an AI bubble today, but it’s sitting on top of a real, long-term revolution in intelligence.
What do you think? Are we in an AI bubble, or just at the beginning of the next industrial age?
#AI #ArtificialIntelligence #Innovation #Leadership #DeepLearning #MachineLearning #TechTrends