The significance of secure data in the AI revolution

In 2026, the conversation around artificial intelligence within the enterprise has fundamentally changed. Moving beyond the need for investment in AI, the more pressing question is how quickly they can scale data and AI capabilities before competitors establish a structural advantage. With speed and execution having become just as important as intent, what was once considered a back-end capability is now being recognised as a driver of enterprise-wide transformation with visible balance sheet impact.

This shift is visible across every function, including sales operations, customer support, risk management, and beyond. Embedded into daily workflows, AI is no longer just a standalone productivity tool but a core pillar of enterprise strategy. However, an often-underestimated enabler beneath every successful AI initiative is data. Enterprises that understand this are not just deploying AI faster but building conditions for it to succeed.

Building a Cogent Data Foundation

The scale of this understanding is reflected in investment patterns. For every dollar invested in AI, up to four dollars are being channelled into data, reinforcing the idea that sustainable AI outcomes are ultimately built on strong data foundations. Despite such investment, many organisations continue to underperform due to entirely skipping the foundational step: the lack of a robust and scalable data landscape to support them. Ultimately, this does not create value but increases risks.

Most organisations already possess a gold mine of data covering operational activities, employee productivity, client behaviour, and market performance. The challenge is not the absence of data but the lack of structure, accessibility, and trust around that data. Holding a substantial amount of data does not automatically translate into usable intelligence that can support critical decision-making. As a result, many companies fail to use their data to its maximum potential.

Another persistent challenge is legacy infrastructure. Approximately 30-40% of enterprise technology spend remains tied to legacy systems, which continues to limit agility and slows down the ability to respond to change. These environments, built over decades, restrict the pace at which AI can be meaningfully scaled. Modernisation is therefore not just a technology upgrade but an organisational imperative.

To build a data foundation that can genuinely power AI at scale, organisations must anchor their efforts on three fundamental success factors.

First, strategy alignment: the AI strategy must be tightly coupled with the business strategy. If an organisation is communicating a boost in profitability or new opportunities with shareholders, its AI roadmap must reflect those same priorities.

Second, business use cases: the focus must be on identifying high-value use cases and creating accelerated pathways for moving from proof of concept to real-world solutions that deliver measurable value.

Third, and most critically, data as the foundation: without a cogent, reliable, and well-governed agentic data foundation, even the most compelling use cases will fail to reach their potential.

Where business outcomes are visible

Organisations that consciously invest in their data foundations are beginning to see tangible results. The immediate and strongest outcomes are emerging in revenue growth, particularly through cross-sell and upsell opportunities. Data and AI are unlocking new revenue through sharper customer insight, improved demand visibility, and the identification of entirely new business models— and this is where organizations see the most immediate strategic value.

Operational improvements are equally significant. In areas such as contact centres, AI-led models are already managing a significant share of customer interactions, improving both efficiency and satisfaction. In some cases, this has enabled up to 50% cost optimisation alongside demonstrably better outcomes. These are not incremental gains but a fundamental rethinking of how enterprise functions operate.

Data-driven & AI led enterprises will enjoy exponential compounding advantages: faster and more confident decision-making, elimination of costly operational errors, reduced overall costs, and the ability to harness the full potential of generative AI without the distortion introduced by poor-quality inputs. The common thread across all these benefits is trusted, accessible, and well-governed data.

From modernisation to competitive differentiation

Many organisations are investing heavily in data modernisation. However, not all of them bring competitive advantages. A significant portion of today’s modernisation is driven primarily by cost optimisation, particularly the shift away from legacy platforms. While necessary, this is not sufficient to create differentiation. What creates true competitive differentiation is the extent to which modernisation is aligned to the core drivers of the business.

For consumer-facing organisations, this often includes modernising customer data ecosystems to enable more effective cross-sell, upsell, and personalised engagement. For others, it may mean building real-time operational intelligence or enabling supply chain visibility. In all cases, the organisations that emerge as genuine winners are those that use data modernisation not just to improve operations, but to fundamentally strengthen how they compete within their industry.

To achieve this, organisations must take a structured approach to transforming their data capabilities. This includes accelerating decision-making by ensuring expedient data analytics across enterprise touchpoints; syndication of data across the enterprise leveraging modern data management tools; leveraging governance frameworks that drive data & AI confidence at every level; and prioritising a human-machine partnership that balances efficiency with accountability. Above all, achieving data excellence requires a culture in which data assets are truly valued and actively used—not simply stored.

Governance, Trust, and the Rise of Agentic AI

As AI systems become more autonomous, trust and governance take the centre stage in enterprise adoption. AI inherently amplifies the quality of the data it is built on: well-structured, contextual and high-quality data leads to better outcomes, while poor-quality or biased inputs can scale risk just as quickly. This makes data quality, representativeness, and bias control critical to how organisations approach AI at scale.

This reality is already reshaping investment priorities. A growing share of AI spend is being directed towards governance, compliance, and data management alongside model development. Organisations are recognising that responsible AI deployment and effective AI deployment are not competing goals, but the same thing pursued through the same means.

Responsible AI is the operating framework for this era. It deals with using AI technology fairly, ethically, and responsibly, establishing guardrails to ensure AI is applied for practical, ethical purposes. Since the right data is central to how AI learns and makes decisions, data governance is the cornerstone of any responsible AI program. This includes addressing bias to ensure that AI systems do not access inappropriate or unrepresentative data that distorts decision-making, violates commitments to fairness, or erodes trust in AI’s reliability.

Responsible AI is not a one-time compliance exercise but an ongoing process that requires regular monitoring of AI system outputs, continuous refinement of data and algorithms where needed, and a willingness to adapt to evolving social norms and regulatory requirements.

What is emerging is a more structured operating model, where AI systems function within clearly defined guardrails, and human oversight focuses on validation, exception handling, and accountability. The ability to balance autonomy with control will ultimately determine the effective scaling of agentic AI in practice by organisations.

Why Data Security Remains Non-Negotiable

As AI systems grow more capable and more deeply embedded in enterprise operations, the question of data security becomes existential. AI models must only be permitted to access data required for their specific operations which have appropriate authorisation. Unprotected data in an AI environment is an expensive gamble that no organization can afford.

The risks are well-documented and severe, with malicious actors increasingly using AI to orchestrate targeted attacks. Data breaches trigger heavy regulatory fines, complex legal exposure, and costly recovery operations. Even inadvertent breaches carry devastating reputational consequences, and exposed data creates a second-order threat with identity theft or manipulation by AI systems that misuse personal or professional information.

Security and privacy therefore are not peripheral concerns. They are pivotal to protecting sensitive information, maintaining regulatory compliance, and preserving the confidence of clients, employees, and partners. Alongside data quality, these elements determine whether an AI system is genuinely trustworthy or merely powerful, an imperative distinction at scale. A holistic governance strategy that addresses quality, security, and privacy in an integrated manner is the only credible foundation for responsible AI adoption.

The Future Belongs to Data-led Enterprises

Looking ahead, the most successful data-led enterprises over the next three to five years will be defined not just by how much they invest in AI, but by how intentionally they build and scale. Several characteristics will distinguish the leaders from the laggards.

They will sustain investment in strong data foundations, prioritising data management, governance, and modernisation, and recognising that scaling AI meaningfully is possible only when the underlying data is reliable and accessible. They will move beyond experimentation, using agentic AI to redesign end-to-end processes and fundamentally rethink how work gets done. They will democratize AI across every function, raising overall capability and making data-driven decision-making part of everyday operations. They will measure success on a larger

horizon, building long-term capabilities that drive compounding value over time. And the most far-sighted among them will extend adoption beyond their own walls, enabling partners and suppliers to participate in their data and AI ecosystems, and creating a network effect that not only reinforces but also amplifies their competitive position.

Today, data is no longer a back-office function. It has become a core business capability that determines the eventual winners in the corporate world. The transition from fragmented data operations to integrated, AI-powered growth engines is not a future ambition; it is a present imperative. The organizations making that transition at scale, deliberately, and securely, are the ones that will define this era. The time to act is now.

(The author is Global Head of Data & Analytics at NTT DATA, Inc. Views are personal)