The AI bubble is no longer a distant warning. It is already taking shape. But for Claude Smadja, the real story is not the flood of capital pouring into artificial intelligence—it is what that flood reveals about a world undergoing a deeper transformation.
Smadja, Co-Founder, India Global Innovation Connect (IGIC) and Chairman of Smadja & Smadja Strategic Advisory, argues that the post-World War II order has already fractured. AI's meteoric rise is unfolding against a backdrop of geopolitical realignment, economic fragmentation, and shifting centres of power.
In that context, the critical question is not whether investment is flowing into AI, but where it is flowing, who controls it, and whether it is creating genuine productivity or merely inflating the next great bubble.
While speaking at AIM Front Page, Smadja turned the signal red for the AI bubble.
"The impact of AI is very, very diffuse," he said, drawing a distinction between individual, corporate, and national-level outcomes.
At the individual level, the benefits are undeniable. "The AI impact on personal productivity exists. It is significant," he said.
Inside companies, however, the results are far more uneven. Depending on the industry, use case, and an organisation's readiness to deploy the technology, productivity gains range from negligible to around 12-15%.
At the level of the broader economy, Smadja is even more sceptical. "The impact on productivity, so far as it can be measured, is just zero for the moment."
For him, that disconnect is precisely what makes the AI bubble a serious concern.
"I do believe that there is a bubble," he said.
The bubble is not merely reflected in soaring valuations or lofty expectations. It is visible in the unprecedented scale of investment flowing into the AI ecosystem—from semiconductor fabs and hyperscale data centres to increasingly ambitious projects, including proposals for data centres in space.
The central question, he argues, is whether this wave of spending will ultimately translate into measurable economic value or is racing ahead of the productivity gains needed to justify it.
Maybe American investor Michael Burry was right all along.
The World War II Theory
Smadja’s warning is not that AI will vanish. It is possible that the current round of spending may be outpacing what the economy can actually absorb. He framed it as a race shaped by uncertainty, over investment, and a growing belief that AI may become a new instrument of supremacy.
That's why he said the bubble “could burst at a given moment," not that it will. The danger isn't just disappointment but the market confusing momentum with proof.
Smadja said the world is already moving through “a new international global architecture” because the old one has become obsolete. The system that emerged after World War II no longer fits the balance of power, the technology landscape or the cultural mood of the moment.
He described the present as a time of “tectonic plates” in motion, with institutions, forces and mindsets all in transition. That creates nervousness, weaker growth and a climate in which people are constantly trying to anticipate what might hit them next.
Smadja views the current shift not as deglobalisation, but as a new, more contested form of globalisation. Unlike the old version—often equated with Americanisation—this iteration is shaped by a shifting balance of economic, technological, and military power.
It also includes stronger regionalisation, which he says was underestimated for years. “It is by far premature to announce the death of globalisation,” he said.
AI is central to that shift. Smadja argued that the rapid rise of frontier technologies has made geopolitics and innovation inseparable. Technology is now a “key strategic national asset,” he said, which is why countries try to protect any lead they have. Full decoupling, in his view, is neither possible nor desirable, even between the US and China. India, he said, is doing something similar in its own way, pushing technological indigenisation while still needing collaboration and cross-border synergies to boost innovation.
No country, he argued, can hide inside a “small bubble.”
That same logic, he said, is shaping the next international order. Rather than a hard bipolar world, he sees “a much more fluid structure” with two broad economic spheres, one organised around China and one around the United States. Countries will gravitate toward one or the other, but most will still maintain ties to both camps.
India in a Better Place
In that setting, Smadja believes India has unusual room to manoeuvre. If it plays its cards right across technology, foreign policy and economic policy, he said, it has the potential to enjoy the maximum possible “sovereign autonomy” in a complex world.
He was strikingly positive about India’s direction. India, he said, has developed a new mindset of entrepreneurship and technological adoption, and its startup story has become global. He also praised the country’s foreign policy for not aligning with any one bloc or superpower.
On manufacturing, he called the “Make in India” push “right on target” because no country can develop fully without a broad industrial base. Supply chains, he said, are not just about assembly. They are ecosystems of firms, skills and efficiencies, and China’s strength has been in building the world’s most efficient ones.
But he also pointed to weaknesses that India still has to fix. Corporate India, he said, needs to do more on research and development. Education and skills remain another major gap.
Human capital, in his view, becomes more important, not less, as AI, quantum computing and semiconductors advance. He argued for stronger apprenticeship-style pathways, closer to the models seen in Germany, Switzerland and Sweden, where young people blend theory with practical corporate training early on.
Without that, he said, a country risks wasting talent instead of compounding it.
He was also honest about why foreign investment has become harder to attract. He pointed to geopolitical volatility, weaker growth, bureaucracy, retroactive tax issues and the fact that some companies walk away after announcing projects.
Even so, he said he is not ready to pass judgment on India’s AI ambitions too early. The country does not have the same financial firepower as the US or China, so it had to take a different approach. In his reading, the government has had to compensate for private-sector underinvestment in R&D, and that is part of why the AI strategy looks different from the Western model.
Still, he thinks India is on the right path. “At this stage, I would say India is playing its cards right,” he said. The real challenge is speed. The country has reached a threshold, with the “roller coaster of despair and hope” now behind it, but the rest of the world is moving too.
The key, he argued, is acceleration through partnerships. Done well, alliances should create a multiplier effect where “1 + one makes three and not just two.” That, more than any single policy slogan, may be the clearest line running through the interview. Smadja’s message is that India’s future, like AI itself, will be decided less by hype than by execution.
Your reaction