Chi-Hua Chien, a venture capitalist with more than twenty years of experience, saw the rise of Facebook coming. Now he's making an equally disruptive prediction: the real winners of the AI era won't be the companies that sell artificial intelligence. They'll be the ones that use it to build concrete value in their own industries.
This view radically changes strategic priorities. It's no longer about picking the best language model or the most expensive platform. It's about understanding where AI can cut operating costs, speed up sales cycles or improve the customer experience. What's more, Italian SMEs, often left out of the tech debate, find themselves in an unexpectedly favorable position: they can adopt tools that are already mature without bearing the R&D costs that weigh on the big players.
In this article, we at SHM Studio break down Chien's thesis, read it against the available data and turn its implications into practical actions for Italian B2B and retail companies that want to position themselves well for the next competitive cycle.
The thesis that flips the dominant AI narrative
For months, the public debate on artificial intelligence has revolved around one clear idea: whoever builds the best models will win the market. OpenAI, Anthropic, Google DeepMind: the narrative rewards those who produce the technology. Chi-Hua Chien, however, offers a radically different perspective.
Chien is a venture capitalist with more than twenty years in the business. He recognized Facebook's potential before it became a global phenomenon. Today, in an interview published on TechCrunch, he argues that the real winners of the AI era won't sell artificial intelligence. They'll use it to create value in their own vertical markets.
This distinction isn't semantic. It's strategic. That's why it deserves a close look, especially for Italian SMEs that have to decide where to allocate resources over the next 18-24 months.
The numbers that cut the AI vendor hype down to size
The AI model market is concentrated in a few hands. According to Gartner estimates, the competition among foundation models is quickly turning into a commodity war. Margins are shrinking. Training costs remain astronomical.
In fact, the paradox is already visible: companies that sell AI as their main product struggle to monetize it sustainably. By contrast, companies like Klarna or Duolingo, which use AI to optimize internal processes, have reported measurable efficiency gains that they can clearly communicate to investors.
Moreover, an analysis by Harvard Business Review confirms that the value of AI is concentrated in vertical applications. Not in generic models. This is key to understanding where competitive advantage will shift in the coming years.
Strategic takeaway: the late-mover advantage
Chien's thesis has a consequence that's often overlooked. Companies that haven't invested in AI R&D over the past three years aren't behind. They're in a position of selective advantage.
They can access tools that are already mature, tested and getting cheaper to adopt. As a result, the opportunity cost of waiting has dropped sharply. What took a team of data scientists in 2023 can now be set up in a few weeks with no-code or low-code tools.
So the right question isn't
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.
