As artificial intelligence (AI) moves from widespread adoption to deeper integration across industries, Athar “Naqi” Naqi keeps in mind the prescient words of software engineer Grady Booch: “A fool with a tool is still a fool.” Naqi, Senior Vice President of Sales and Client Success at ekSource Technologies, Inc., sees a growing problem in the wave of AI initiatives emerging across industries. As organizations face increasing pressure to move quickly, often before establishing the governance and operational readiness needed to support long-term success, many initiatives are failing to deliver measurable business value with AI. “Buying AI without a strategy doesn’t make you innovative, it makes you an expensive cautionary tale,” says Naqi. One of the biggest reasons for this, he says, is a persistent misconception that AI behaves like traditional enterprise software.
Why Enterprise AI Initiatives Fail
“The biggest mistake is treating AI like a software purchase rather than a strategic transformation,” he says. “Without a clear problem statement tied to your business outcomes, you end up with impressive demos but disappointing ROI.” Unlike enterprise resource planning (ERP) or customer relationship management (CRM) deployments, AI systems continuously evolve and influence how employees make decisions. Organizations focused solely on implementation speed often overlook the people and process changes required to sustain AI return on investment (ROI) over time.
Naqi points to organizational culture as a deciding factor in whether AI strategy succeeds or collapses. “Bending Peter Drucker’s words here: culture will absolutely devour AI implementation for lunch.” The companies seeing meaningful returns are treating AI readiness as a leadership issue rather than an IT project. They invest in communication, training, governance, and realistic timelines alongside the technology itself. They also avoid delegating accountability too early in the process. “The ‘burned’ leaders delegate accountability too early,” Naqi says. “The successful ones stay personally engaged through the implementation, and ownership at the top is non-negotiable.”
AI Governance Cannot Be Delegated
As AI systems gain more autonomy, governance is becoming one of the defining responsibilities of the modern C-suite. Questions around AI accountability, AI risk, and regulatory compliance are increasingly tied to board-level oversight rather than technical administration. “AI decisions are fundamentally decisions about risk, liability, brand reputation, and competitive positioning,” Naqi says. “These have always been boardroom responsibilities.”
This shift is already being accelerated by emerging regulatory frameworks such as the EU AI Act (2024) and evolving U.S. governance proposals. Executives are being forced to rethink how they evaluate AI vendors, measure accountability, and establish internal controls before scaling deployments. “Technical teams can tell you what AI can do,” he says. “Only the C-suite can determine what it should do.” This is important as organizations adopt generative AI coding assistants and autonomous workflows. Faster code generation may improve productivity, but it does not automatically guarantee operational stability or production readiness. Without oversight, organizations risk introducing vulnerabilities, inefficiencies, and governance gaps at scale.
Building Circuit Breakers Before Scaling AI
Naqi stresses that governance infrastructure must come before deployment, pointing to the May 6, 2010 Flash Crash. Autonomous trading algorithms erased nearly $1 trillion in market value in less than 36 minutes before recovery mechanisms stabilized the market. “We already ran this experiment with autonomous agents in financial markets,” Naqi says. “Before you deploy agents at scale, you have to ask yourself, do you have circuit breakers?” Autonomous systems can create autonomous failures. As companies move toward agentic enterprises, where AI systems negotiate, analyze, and execute decisions independently, the need for oversight becomes significantly more urgent.
Naqi advocates for a measured AI decision framework for C-suite leaders centered around phased deployment, human oversight, and escalation protocols. Organizations should start narrow, instrument everything, and establish governance before allowing AI systems to operate independently in high-stakes environments. “Governance infrastructure is not a phase two,” he says. “It should be phase one of your project.” Executives must also define exactly when AI systems hand decisions back to humans and who ultimately owns accountability.
Separating AI Hype From Business Value
The conversation around AI strategy is shifting from productivity tools to operational transformation, but many executives are still focused on outdated AI use cases. “Most executives are still skating to where AI was in 2023.” The next phase of enterprise AI will be defined by organizations capable of combining proprietary data strategy with AI-literate leadership teams. Companies that can interrogate AI outputs instead of blindly trusting them will have a significant advantage as AI systems become embedded deeper into business operations.
As the focus moves away from productivity gains toward long-term operational resilience, the questions C-suite leaders must ask before buying AI need to become more sophisticated. Executives who understand how to measure AI ROI before scaling are far more likely to avoid costly AI investments that fail under real-world conditions. Naqi offers a simpler framework for navigating this shift. Measure twice before cutting once. “AI is not a prototype anymore,” he says. “Organizations must change themselves to adopt AI, not just purchase technology.”
Follow Athar “Naqi” Naqi on LinkedIn for more insights.
