Leading in the Age of AI: Prioritize Problem-Solving Over AI Adoption Hype

When you are around the IT block for a few years, you get to see your share of tech revolutions— cloud migrations, blockchain buzz, crypto magic, etc. But nothing matches the frenetic pace of AI that we are witnessing today. The weekly announcement of the latest and greatest Large Language Model is no longer a surprise or earthshaking news unless it is a DeepSeek-level calamity that wakes everyone all at once. Startups flooding the market with a tool promising to revolutionize everything is now as commonplace as the next dotcom company that came around in the late 90s. All this while, companies are racing to build their “own” platforms in the name of domain specificity, security, or dominance-related considerations.

The pressure is palpable: from peers flaunting their AI subscriptions, to boards demanding “AI strategies,” leaders feel compelled to act—often without pausing to ask “why”. The compulsive need to demonstrate that they are entirely up to speed with AI and are actively embracing it is bordering on levels that are neither healthy nor sustainable. Not touting your AI adoption strategy is as close to career suicide for individuals and becoming archaic for a company – not great options in today’s volatile economy.

Between signing up for ChatGPT Enterprise, licensing platforms like Hugging Face, or greenlighting custom AI builds, companies are starting to realize that they’ve adopted AI solutions in search of real problems for their customers.

From AI having to be a strategic leadership imperative—it’s now become peer-driven panic amongst equals – for individuals and companies alike. In Fintech, where security, speed, scale, and customer trust are non-negotiable, we cannot afford to let hypes dictate our moves. Instead, we must anchor ourselves in timeless principles that have been the bedrock of our industry: solve real constraints today, preempt bottlenecks tomorrow, and enhance service for customers—internal and external.

The AI Adoption Trap: Peer Pressure Over Purpose

The AI race has turned into a status game. Truly so. At a recent industry conference, a fellow CTO boasted about deploying Google’s Vertex AI across his organization—a $100,000 monthly subscription, thousands of employees trained, yet no clear metric on how it improved loan processing times or fraud detection rates. Another exec I know pushed his team to build a proprietary LLM, citing “competitive edge,” only to burn $2 million and six months before shelving it for lack of a use case. These aren’t outliers. A 2024 McKinsey survey found that 65% of executives feel pressured to adopt AI, yet only 25% report meaningful ROI from their investments. The disconnect? Leaders are buying tools—Vertex AI, AWS SageMaker, OpenAI’s API—or building platforms because others are, not because they address a specific pain point. The self-imposed compulsion to look cool, while understandable, is a toxic behavior pattern that needs to be recognized, stopped, and approached from a rational and customer-centric perspective.

This behavior mirrors the Fintech bubble of the mid-2010s when blockchain was the shiny object. Companies poured resources into proofs-of-concept, only to find most didn’t scale beyond pilots or there weren’t use cases that made sense for their immediate adoption. AI’s allure is even stronger, fueled by endless vendor pitches and startup noise – not to mention the rapid proliferation of LLM versions and the continuous introduction of features and functions that will somehow need to be consumed before the hype is called out. The result: a cluttered market where leaders grab subscriptions or tools to signal “innovation,” sidelining the harder work of identifying what’s broken or bottlenecked in their operations.

Refocusing on Constraints: The Leadership Imperative

Great leaders don’t chase trends; they solve problems. That is the bottom line. Constraints abound in Fintech: legacy systems slowing transaction speeds, manual compliance checks delaying customer onboarding, the delay in implementation schedule, or fighting the eternally vicious risk-and-compliance battle. Future bottlenecks loom, too—think regulatory shifts demanding real-time reporting or scaling customer support as digital banking explodes. And customers, internal and external, crave better—faster resolutions from support teams, and seamless experiences for account holders. AI can address these elegantly, but only if we start with the problem, not the solution.

Leadership Behavior: Curiosity, Not Conformity

Adopting AI wisely demands a shift in how we lead our teams and companies. Too often, I see execs deferring to vendors or mimicking competitors—signing up for OpenAI’s enterprise tier because “everyone else is.” That’s conformity, not strategy.

IFFCO-Tokio, an Indian insurer, tackled fraudulent motor claims eating into profits. Rather than chasing broad AI hype, their CIO, Seema Gaur, leveraged the H2O AI Cloud to analyze claims data, spotting anomalies with pre-built algorithms. Deployed on-premise in July 2021, it saved over $1 million annually (IFFCO-Tokio Saves Over $1M Annually on Fraud). A 2023 Deloitte study backs this fact: firms with “problem-driven” AI strategies see 3x higher ROI than those chasing “tool-first” adoption. Talk about the proof being in the pudding – this is precisely it.

The Path Forward: Purpose Over Pressure

The AI frenzy won’t slow—models will get faster, tools shinier, and peer pressure louder. But leadership isn’t about keeping up; it’s about cutting through. In Fintech, where margins are tight, and trust is king, we can’t squander resources on solutions without problems. Start with your constraints—legacy drag, scaling pains, service gaps—then find or build what fits. The CTO who solves a $1 million bottleneck with a $10,000 tool outshines the one who spends $1 million to signal “AI-forward.”

So, resist the subscription spree. Skip the “me too” builds. Focus on what’s real, now and next. That’s how we turn AI from a buzzword into a business win—and lead with purpose, not panic.