The Complexity Tax: Are Organisations Caught Up in the AI Adoption Wave?

New Delhi, India Jul 8, 2026 ALN: In the digital world, every groundbreaking piece of technology is perched atop something that already exists. This layering of technologies creates a complex ecosystem where new innovations are built upon older systems, which can lead to both opportunities and challenges. For instance, the Internet is built over client-server architectures; Software as a Service (SaaS) is constructed above on-premises procurement cycles; and Cloud solutions are deployed over security models that were originally designed for data centres. Currently, Artificial Intelligence (AI) is stacked above them all, resting on a wobbly foundation that can hardly withstand the added weight of its own complexities.

These patterns suggest a fundamental truth—the issue is never really with the technology that is introduced; it is with the complexity that lies deep within the existing systems and processes. As organisations rush to embrace AI, they often overlook the intricacies of their current infrastructure, leading to what has been termed the "Complexity Tax." This phenomenon highlights the hidden overhead costs that businesses incur as a result of fragmented systems, processes, and shadow technologies.

Complexity Tax—What it Means and How it Accumulates

According to BCG’s AI at Work 2026 Report, India takes the global lead in workplace AI adoption, with an impressive 74% of frontline employees actively using AI tools. This trend is mirrored in the mid-market segment, as highlighted by the recent Cost of Complexity 2026 report. It reveals that India tops the global mid-market AI integration, with 36% of organisations embedding AI across multiple core operations, which is more than double the 15% global average.

Organisations adopt AI with the expectation that it will enhance critical business functions, ranging from customer support and human resources to finance and IT, ultimately bringing these disparate areas together. However, the reality often proves to be painstakingly different. Many organisations have resorted to various platforms to cater to their immediate and varied business needs, resulting in systems that operate effectively in isolation but struggle to work in unison. This fragmented approach leads to operational inefficiencies and contributes to the complexity tax.

Stacking AI over this fragmented foundation exposes latent operational gaps and introduces errors and delays that are already present in the IT environment. It is crucial to note that AI is not to blame for these issues; rather, it is the underlying complexity of existing systems that creates these challenges. As organisations increasingly integrate AI into their operations, they must confront the reality of their complexity tax.

The Consequences and the Solutions at Hand

The complexity tax has significant consequences for organisations. According to the Cost of Complexity 2026 Report, mid-market organisations lose 25% of their AI spend due to the overhead operational costs associated with managing these complexities. Furthermore, 88% of Indian mid-market IT leaders report that managing AI complexities is increasing their teams' workloads. This indicates that the damages caused by complexity are not merely impacting budgets; they are also threatening overall productivity.

Research by McKinsey highlights that nearly two-thirds of organisations have not yet begun to scale AI across the enterprise. Despite the potential for innovation, as indicated by 64% of respondents who believe AI is enabling innovation, only 39% report a positive impact on earnings before interest and taxes (EBIT). This disparity underscores the challenges organisations face in effectively harnessing AI's capabilities while navigating the complexities of their existing systems.

Fortunately, the situation is not beyond repair. The key to overcoming the complexity tax lies not in the technology itself, but in the structural changes that organisations must implement to support successful AI integration. Here are four critical structural changes that must take place to ensure that organisations thrive amidst the rapid AI adoption:

• Execution and experimentation must happen together: As new waves of technology emerge, organisations often feel compelled to experiment. However, experimentation should never come at the expense of execution. Instead, these processes should occur in tandem. Organisations must establish clear criteria and metrics to facilitate the transition from experimental pilot stages to final executive production, ensuring that both innovation and operational efficiency are prioritized.
• Resolve fragmentation through integration: Fragmented data, tools, and processes inherently lead to complexities within organisations. To mitigate this issue, it is essential that organisations thoroughly evaluate inbound AI tools to ensure they can be seamlessly integrated into the necessary business functions. By doing so, organisations can create a more cohesive operational environment that reduces the complexity tax.
• Choose prioritisation over customisation: While it may seem logical to customize tools to meet rising business needs, organisations should prioritize selecting tools that offer a commitment to time-to-first-value. This approach allows organisations to minimize complexity costs while still addressing their operational requirements effectively.
• Focus on outcomes, not capabilities: The success of any AI initiative should be measured by the business value it generates rather than the technological capabilities it offers. Organisations must track improvements achieved through AI, such as quicker workflows, reduced manual effort, or enhanced customer experiences. This focus on tangible outcomes will help maintain stakeholder confidence and ensure that AI investments remain aligned with real operational results.

Every era brings with it the promise of a new phase of digital transformation, and each generation is drawn to the possibilities these advancements present. However, organisations must recognize that real value is derived not just from the technology itself, but also from the foundational structures upon which it is introduced. As AI adoption continues to soar, organisations that neglect to address their structural challenges and cover their complexity tax risk losing valuable time and resources managing fragmented systems. They may miss out on performing tasks that yield significant business outcomes.

Ultimately, an organisation’s success is no longer solely dependent on its ability to adopt AI; it hinges on how effectively they build the necessary structures to support it. As the landscape of technology continues to evolve, those organisations that can navigate the complexities inherent in their systems will be better positioned to leverage AI for sustainable growth and innovation.

— Murali Swaminathan, Chief Technology Officer, Freshworks