Mizuho’s CDAO: Stop Doing AI Just for the Sake of AI

In today’s fervent tech landscape, where Artificial Intelligence is on everyone’s lips, many voices praise its unlimited potential. But when I read the statement by Mark Wilden, CDAO (Chief Data & Analytics Officer) of Mizuho Americas, I was immediately struck by his candor and realism. His assertion that ‘doing AI for the sake of AI destroys value’ resonated deeply with me, striking a raw nerve that, in my daily work with ‘Cortex News’, I see ignored far too often: the obsession with technology for its own sake.

The Risk of AI for AI’s Sake

Wilden is not the first to issue this warning, but his position within a financial giant like Mizuho gives his words special weight. According to him, the problem is not AI itself, but the superficial approach with which many companies adopt it. They invest in expensive machine learning projects or LLM-based solutions simply because ‘we need to have AI’, without first identifying a concrete problem to solve or a tangible value to generate. This mindset leads to wasted resources, disillusionment, and ultimately a negative perception of a technology that, when properly applied, can truly be revolutionary.

I have personally seen how initial enthusiasm for a new AI implementation clashes with the harsh reality of poorly defined or absent success metrics. Without a clear connection to business goals—whether improving operational efficiency, optimizing the customer experience, or creating new revenue streams—an AI project becomes a pure cost, a technological ‘showcase’ that leads to nothing concrete. It is like buying the most powerful car on the market without knowing where you want to go, driving aimlessly and burning precious fuel.

From High Expectations to Real Value: Building a Solid AI Strategy

The key, as Wilden suggests and as I have always maintained, lies in flipping the perspective. You don’t start with AI; you start with the business problem or opportunity. ‘What is the pain point we want to alleviate? What is the process we can drastically improve? What is the new service we could offer?’ Only after answering these questions should one evaluate whether Artificial Intelligence is the most suitable tool to achieve those goals. This is the foundation of a truly value-driven ‘AI strategy’.

Implementing AI requires a deep understanding of available data, infrastructure capabilities, and, above all, internal skills. It is not a light switch you flip and it works on its own. It requires planning, iteration, and the ability to rigorously measure impact. Mizuho, being an organization with massive volumes of data, understands that every decision must be data-driven and, in the context of AI, value-driven. The risk is investing millions in complex models that no one knows how to use or, worse, that produce results misaligned with business needs, generating frustration and skepticism.

From an engineering perspective, Mark Wilden’s provocation is an essential wake-up call for the entire industry. We cannot afford to fall into the trap of innovation for innovation’s sake. The true value of Artificial Intelligence only manifests when it serves a higher purpose, integrated into a clear and measurable strategic vision. It is easy to get carried away by the hype of new technologies, but a true innovator knows how to distinguish signal from noise, always focusing on tangible value for the company and users. What do you think, dear readers? Have you ever seen AI projects fail precisely due to a lack of a clear value vision, or conversely, what examples of well-applied AI have struck you the most?