NEW DELHI — As global enterprises scramble to integrate cutting-edge artificial intelligence into their core operations, a formidable roadblock is quietly emerging from within: legacy tech infrastructure. According to Kulmeet Bawa, Managing Director and Global Vice President (GVP) for India and SAARC at ServiceNow, modern organizations are heavily siloed, leaving their underlying datasets and workflows ill-prepared for the seismic shift of an AI-driven economy.
Speaking exclusively to businessline during its premier flagship event in India, Bawa dissected the profound friction points businesses face today. From the compounding complexities of mergers and acquisitions to the stringent demands of data sovereignty and regulatory compliance, Bawa emphasized that unleashing generative AI onto fragmented systems is a recipe for operational chaos.
As financial markets navigated a turbulent session—with the Sensex dropping 429.11 points to close at 72,638.70, and the Nifty shedding 173.05 points to settle at 22,603.05—the macroeconomic backdrop underscored the urgency for corporate resilience. In an era where efficiency, agility, and risk management dictate market survival, fixing the structural plumbing of enterprise IT is no longer optional; it is an existential imperative.
1. Main Facts: The AI Integration Paradox
The core thesis presented by ServiceNow’s regional leadership points to a glaring contradiction in modern enterprise strategy: while executive boards are universally obsessed with deploying AI agents and large language models (LLMs), they are simultaneously sitting atop decades of fragmented, siloed, and aging IT infrastructure.
- The Silo Trap: Over the past 15 to 20 years, companies have meticulously built disparate "systems of record" across individual departments—HR, finance, customer support, and supply chain—without creating unified enterprise workflows.
- The M&A Multiplier: Mergers and acquisitions exacerbate this architectural debt. When two legacy enterprises combine, their disparate data lakes and workflow mechanisms rarely integrate cleanly, creating a breeding ground for operational friction.
- AI as an Accelerator of Chaos: Bawa warns that introducing advanced automation and AI into a fragmented ecosystem will not streamline operations; rather, it will amplify existing inefficiencies, scaling errors and hallucinations at unprecedented speeds.
- The Regulatory Shield: Driven by strict local frameworks—including mandates from the Reserve Bank of India (RBI)—enterprises are waking up to the reality that compliance and data sovereignty cannot be "bolted on" as an afterthought. Guardrails must be woven into the foundational architecture from day one.
2. Chronology: The Evolution of Enterprise IT and the AI Rush
To understand how global corporations arrived at the current infrastructural bottleneck, one must look at the historical trajectory of enterprise software development over the last two decades.
Phase 1: The Era of Siloed Systems of Record (2000s – 2010s)
For the better part of twenty years, the primary goal of corporate IT was digitization. Organizations rushed to deploy specialized software for individual business units. Customer relationship management (CRM) tools lived in isolation from enterprise resource planning (ERP) systems, which in turn had little to no native visibility into human resources or operational technology (OT). While these platforms successfully recorded digital transactions, they functioned as standalone islands.
Phase 2: The M&A Wave and Architectural Complexity (2015 – 2023)
As global markets consolidated, high-frequency mergers and acquisitions became the norm, particularly in the IT, telecommunications, and financial services sectors. Companies routinely absorbed competitors or acquired niche technology startups. However, integration strategies historically focused on financial consolidation rather than technological harmonization, leaving surviving entities with a tangled web of legacy databases, redundant applications, and fractured workflows.
Phase 3: The Generative AI Gold Rush and the Awakening (2023 – Present)
Following the explosive mainstream debut of generative AI, the corporate world entered a phase of unbridled experimentation. Initially, C-suite executives engaged in a race for scale—measuring success by the sheer volume of AI agents deployed and capital expended.
However, by late 2024 and extending into 2026, a sobering realization set in. Executives discovered that uncoordinated AI deployments over dirty, siloed data led to erratic outputs, security vulnerabilities, and compliance failures. This prompted a strategic pivot: the conversation shifted away from reckless spending and toward robust data governance, clean architectures, and unified workflow automation.
3. Supporting Data & Market Context
The challenges outlined by ServiceNow are reflected across broader economic and financial indicators. As corporations reallocate capital toward digital infrastructure and governance, the global markets continue to experience volatility driven by commodity price shifts and shifting monetary policies.
Market Snapshot (As of Reporting)
- BSE Sensex: 72,638.70 (Down 429.11 points)
- NSE Nifty: 22,603.05 (Down 173.05 points)
- Crude Oil: ₹8,815.00 (Up ₹187.00)
- Gold: ₹148,500.00 (Down ₹1,620.00)
- Silver: ₹222,867.00 (Down ₹4,375.00)
Enterprise Tech Spending Trends
According to industry analysts, corporate spending on enterprise workflow modernization and AI governance platforms is projected to grow exponentially over the next three years. Companies are increasingly prioritizing platform-based approaches that sit above legacy systems, acting as an intelligent orchestration layer to harmonize data before it reaches AI models.
Simultaneously, the rise of Global Capability Centers (GCCs) in India has transformed the country into a critical nerve center for global digital transformation. India is no longer viewed merely as a back-office execution hub, but as a strategic incubator for enterprise innovation, workflow re-engineering, and advanced AI control frameworks.
4. Official Responses and Insights: Q&A with Kulmeet Bawa
During his interaction with businessline, Kulmeet Bawa offered candid insights into the realities of digital transformation, client psychology, and ServiceNow’s strategic vision for India and the SAARC region.
Q: There have been many examples of IT companies merging over the past year. What happens to these legacy datasets, and what are the new challenges?
Kulmeet Bawa: The problem of legacy workflows or fragmented workflows will be more pronounced today than ever before. When two companies merge or one acquires the other, this problem gets magnified. Even if they are separate, enough companies are understanding they first need to solve for these fragmented workflows.
What has happened in the last 15-20 years, we’ve consciously built systems of records in every company, but if you look at any enterprise today, it is heavily siloed. If AI hits a fragmented system like this, it will give you chaos. If either of those two companies realise they need to have a cleaned-up, non-fragmented system, they are better off in the AI world.
Q: How have client conversations evolved at a time of hyper-focus on data sovereignty?
Kulmeet Bawa: We are paranoid on data sovereignty because we want to become the governance engine. From an India perspective, we are in deep conversations. We understand the data regulations, what’s happening on sovereignty, RBI expectations.
At first, CXOs were in a race for the most number of agents, who’s spending more. Very soon, they realised that that is a recipe for chaos. So, the guardrails have to come in from the inception level, not bolted on later. That’s a change I’m seeing even in the last six months with customers.
Q: What is your take on the debate around slowing down AI progress, specifically for India?
Kulmeet Bawa: I don’t think it’s about slowing down, but a balance of trust with pace. It’s not an either-or situation. None of us can afford to slow down in today’s world. The last thing we want is a company shutting down because they’ve not been paranoid on building in or seeding in compliance at the very start.
A lot of organisations do not have any visibility on the operational technology, factory flows, medical devices. That’s a nightmare. Imagine where this will lead us in a year. We need to leapfrog; we are a nation that knows how to operate at scale. It’s a learning curve.
Q: What is your plan for India going forward? Do you intend to expand the current workforce?
Kulmeet Bawa: It’s not about the number of people, but the productivity they bring. We’ve got massive headroom to grow in India alone, and if the business demands it, we will absolutely hire.
In terms of priorities, for me it always starts with: how do we forge deep relationships with the largest of enterprises? Then, GCC acceleration is super critical to me, not just because of business in India, but because this can impact India overall. Lastly, they call me a dreamer, but I would love to get India to be the showcase for agentification, innovation, and AI control in the AI world.
"We’ve got massive headroom to grow in India alone, and if the business demands it, we will absolutely hire." — Kulmeet Bawa, MD & GVP, ServiceNow India and SAARC
5. Strategic Implications: The Path Forward for Enterprises
The warnings and insights shared by ServiceNow leadership carry profound implications for enterprise architects, chief technology officers (CTOs), and business leaders navigating the current technological landscape.
1. The Death of the "Bolt-On" Compliance Model
The days of deploying AI models first and figuring out data security and regulatory compliance later are officially over. Regulatory bodies across jurisdictions—particularly in heavily regulated sectors like banking, financial services, and healthcare—are increasing scrutiny on algorithmic transparency and data residency. Enterprises that fail to embed governance guardrails into their foundational system architecture face severe penalties, reputational damage, and operational disruption.
2. The Rise of Unified Orchestration Layers
To overcome the limitations of legacy silos, companies must invest in enterprise platform layers that abstract complexity. Rather than ripping out and replacing expensive legacy ERP and CRM databases—a financially prohibitive and high-risk endeavor—successful organizations are deploying intelligent workflow platforms that sit on top of legacy systems, harmonizing data streams and providing a unified pipe for AI consumption.
3. India’s Pivotal Role as a Global Innovation Sandbox
With its unique concentration of engineering talent, rapidly evolving regulatory framework (such as the Digital Personal Data Protection Act), and booming Global Capability Center ecosystem, India is uniquely positioned to pioneer enterprise AI governance. As Bawa noted, transforming India into a global showcase for "agentification" and responsible AI control could establish a blueprint that other markets emulate.
4. Balancing Velocity with Operational Vigilance
Ultimately, the transition into the AI era requires a delicate calibration between speed and safety. Organizations that mistake reckless deployment velocity for genuine innovation risk structural collapse. By focusing on data cleanliness, workflow integration, and institutional trust from the ground up, enterprises can successfully bridge the chasm between legacy constraints and future-ready autonomy.
