Nvidia’s market capitalization nearing $5 trillion has become the defining symbol of artificial intelligence’s explosive rise, and, according to Nigel Green, CEO of financial advisory group deVere, a clear sign that investor enthusiasm may be running far ahead of profit reality.
“AI is transforming the global economy and remains the single most powerful force shaping the future,” Green said in a statement on Tuesday. “But valuations are expanding faster than earnings. The technology is real, the transformation is real — but the profitability still has to prove itself.”
Nvidia’s extraordinary climb has been powered by soaring demand for its advanced chips that fuel AI systems across industries. Its revenues have more than doubled in the past year, with its data-center business contributing the majority of its record quarterly revenue of more than $44 billion. The company’s dominance has made it the world’s most valuable semiconductor manufacturer and the focal point of the AI revolution sweeping global markets.
Yet Green warns that much of the sector’s growth remains built on anticipation rather than tangible earnings. “These numbers reflect breathtaking momentum, but they also expose the imbalance between demand for capacity and evidence of sustainable profit,” he said. “We’re still waiting to see where the profitability lands once the AI ecosystem matures.”
He added that the current market dynamic is fueled by what he called “expectation layered on expectation,” with major technology firms effectively trading among themselves. “The ecosystem is trading with itself — chipmakers selling to hyperscalers, software firms selling to one another,” Green said. “It looks like growth, but it hasn’t yet generated the profits to justify these numbers.”
According to Green, the AI value chain remains largely circular, with the same companies acting as both the biggest buyers and sellers of AI capacity. “We’re seeing investment feeding on momentum,” he warned. “At some stage, markets will demand evidence that this cycle can sustain itself through actual profit.”
While some investors have expressed concern that the AI rally could be overheating, Green views this phase as a natural evolution. “All revolutions reach a point where the narrative has to meet the numbers,” he said. “We’re approaching that point. Investors will soon want to see whether AI’s extraordinary promise can be matched by earnings that justify the scale of belief.”
Drawing parallels to earlier technology waves, Green noted that industries such as the internet, clean energy, and mobile computing also experienced periods when capital poured in faster than profits appeared. “When the commercial models eventually caught up, they reshaped the global economy,” he said. “AI will follow a similar path — but this time, the sums involved are far larger.”
He also pointed to growing concentration risks in the current market, where a handful of firms dominate chips, software, and cloud infrastructure. “That concentration is driving markets higher, but it also makes them fragile,” he said. “A sustainable AI economy will depend on broader profitability across the ecosystem, not just a few extraordinary valuations.”
Despite cautioning about inflated expectations, Green remains optimistic about AI’s long-term investment potential. “This is not about retreating from the sector. It’s about investing with intelligence,” he said. “The winners will be those using AI to create measurable productivity in energy, logistics, healthcare, and finance — not just those building the infrastructure.”
He concluded that an eventual correction in the sector would be both necessary and constructive. “The coming profit check is healthy. It will clarify who’s delivering real value and who’s still trading on expectation,” Green said. “When that correction comes, it won’t kill the AI story — it will confirm it. The technology is unstoppable, but the valuations are unsustainable without profit to back them. This is the moment for investors to stay engaged, stay selective, and focus on where AI’s promise turns into performance.”
