CIOs Shift Strategies As AI Forces A Balance Between Operational Savings And Innovation

Technology leaders are being pulled into two directions at once. CIOs are expected to use AI to streamline the business and cut costs, yet they're also expected to turn the same technology into new products, new markets, and new revenue. The result is a strategic balancing act where efficiency and innovation compete for the same resources, and the stakes rise with every decision.

This tension is familiar terrain for Aravind Kashyap. As Chief Information Officer at sporting goods manufacturer Riddell, he relies on a decade of experience steering technology toward business growth across multiple executive positions. His past leadership at SAGE IT, INC and Hexaware Technologies informs his view that the real opportunity lies in masterfully combining both mandates rather than choosing one over the other.

"When it comes to using AI for cost reduction or new revenue, the answer is almost always both. Every decision must tie to the organization's growth strategy because the era of chasing shiny ideas is over. The role now is to balance efficiency and innovation in service of where the business is heading," said Kashyap. To strike that balance, he explained that communication must be anchored firmly to business strategy. Kashyap broke his AI decision framework into three distinct lenses to help CIOs decide where AI truly belongs.

• ROI horizon: "The first lens is the return on investment horizon we're expecting. If the pressure is short term, AI for crushing costs or driving operational efficiency becomes important. If the horizon is twelve to twenty four months, then AI for growing sales makes sense because sales has a lead time," he stressed. This timing check keeps teams honest about whether an initiative solves for immediate efficiency or future growth.

• Data maturity: "AI succeeds only if the data is ready. If an organization has strong operational data, it should use AI to minimize costs. If it has rich customer data, it should use AI for growth and revenue opportunities," Kashyap said. This lens shifts the focus from ambition to capability and forces organizations to ground their AI plans in the data they actually have.

• Business priority: "Every company has a priority. If the focus is margin pressure, AI for cost reduction is essential. If the focus is market expansion or new products, then AI for sales growth makes far more sense." Kashyap's final lens ensures AI aligns with the business agenda, not the other way around. "It starts with identifying the real business pain point: cost, noise, or compliance. There's no other way to do it."

Of course, even the best strategy hits roadblocks. Kashyap noted that these challenges often arise when a belief in the myth of "zero marginal cost" AI leads some leaders to underestimate the long-term debt required to govern and secure new systems.

• Data everywhere, data nowhere: Most AI roadblocks start long before the model is built. "Data layer complexity is one big challenge most companies are still addressing because data is all over and nowhere to be seen," explained Kashyap. "Data lineage problems, data quality problems, and data governance problems are all architectural design challenges."

• The running meter: "We often look at cost as an afterthought because many of us do not know the true cost of AI. We think launching a project is easy, but we overlook the long term implications," he noted. "AI workloads run on high performing cloud infrastructure, and the meter keeps running the moment you start using them." This is where many pilots fail to scale, not because the idea is weak but because the ongoing consumption costs were never modeled.

• Guarded and governed: Kashyap pointed to security and compliance as the final major obstacle. "We're still in the early stages of figuring out how to secure AI infrastructure and applications. There's the challenge between centralized security versus decentralized capabilities, like zero-trust pipelines and federated learning. A lot of that is still at a low level of maturity in terms of governance and compliance."

These challenges are not reasons to pause but reasons to begin. Progress comes from movement, not perfection, Kashyap insisted, and the most effective CIOs resist the urge to wait for cleaner data, stronger platforms, or fully mature governance. Momentum creates clarity. "It's a lesson we're all learning, and AI is only accelerating that path. You can wait for five years hoping things will mature, but then you have missed the boat. You will never have a state of nirvana where everything is given to you. You just have to execute," he concluded.