AI is changing the way organizations operate. That much is undeniable.

The push to adopt AI has shifted many executives’ mindsets toward their organizations’ data, highlighting critical gaps in data availability and accessibility. Consequently, over the last few years, an immense amount of money has been invested in back-end data engineering to compensate for years of underinvestment.

This work has begun to break down inter-departmental barriers to data accessibility and availability. The question now becomes how we’ll leverage this investment once the AI value curve plateaus.

This question is particularly interesting in physical industries (energy, rail, utilities, and manufacturing), as these industries sit on mountains of operational data and operate through complex, interconnected systems. These organizations have built their cultures around reliability, uptime, quality, and safety, priorities that have served them well.

But those same priorities can make organizations risk-averse and slow to embrace change, often reinforcing the structural and regulatory barriers that keep data siloed. How do they use the investments they’ve made in their data infrastructure once we reach a post-AI boom world?

The key is adopting end-to-end value chain thinking: moving from siloed, departmental optimization to holistic, system-level collaboration across entire industries and shifting mindsets from siloed data and individual performance to one of global change and systemic improvement, at both a departmental level and at an organisational level.

Let us give you an example of what we mean.

AI Momentum Has Finally Created the Catalyst to Break Silos

Now, the real transformation is cross-organizational and systemic.

Consider a large rail infrastructure owner responsible for tens of thousands of miles of track. Who has the best visibility into the condition of those lines? It’s not the infrastructure owner; it’s the dozens of rail operators running thousands of trains daily across that network.

These operators have proximity to assets in a way the infrastructure owner can’t gain on its own. Natural collaboration between these organizations is clearly mutually beneficial. By using rail operators’ data and insights, the infrastructure owner gets a more detailed and accurate overview of asset quality than it could ever achieve independently. In return, rail operators get better-maintained lines to run on, minimizing disruption and downtime to their services.

This kind of mutual benefit is common sense. So, why does it not happen more regularly?

Physical industries operate critical assets where failures have significant knock-on effects and quickly affect thousands or even millions of people. With rising demand, the pressure to hit sustainability goals, extreme weather disruptions, and increasingly strained infrastructure, organizations can no longer operate in isolation. They must collaborate deeply with partners, customers, and adjacent players in their ecosystem.

Transmission providers have to support energy demand at an exponentially growing rate due to large load centers and extreme weather causing more frequent once-in-a-decade issues (grid strain from heating/cooling demand spikes, infrastructure damage from storms, unpredictable load patterns). In terms of critical physical assets, equipment failure or unexpected spikes can impact millions. Organizations will be forced to embrace end-to-end value chain thinking.

Given the immense value that may be on the table, why wait for disruption to force change? Why not use the momentum gained from the recent AI boom to seize this opportunity now?

When assets are owned by different companies under siloed SLAs, end-to-end thinking isn’t incentivized, but protective thinking is. This mindset cascades through organizations, becoming a blocker for using digital tools to their full extent. Contracting structures reinforce these silos, and regulations haven’t yet evolved to make room for innovation, all within environments already averse to change.

Many organizations are stuck in a cycle: teams focus on protecting SLAs while legacy structures reward isolated performance. Digital tools and technology can act as a spanning bridge, connecting through both inter- and intra-organizational silos. The work being done to provide accessible and available data, motivated by the AI boom, provides a first step in many cases toward overcoming these barriers.

Breaking Down Silos Within and Between Organizations

The momentum created by AI-driven data infrastructure should be used now to break down silos, both within organizations and between them. This means fostering:

• End-to-end operational integration: Connecting maintenance, procurement, inventory, and operations across multiple organizations
• Cross-organization data sharing: Enabling faster anomaly detection, predictive maintenance, and reduced operational failures
• Shared discovery exercises: Identifying insights no single stakeholder could uncover alone

What Could This Look Like in Practice?

Let’s look at energy grid spare-parts management, which today involves disconnected functions (maintenance, construction planning, inventory, and procurement), each working in isolation. This leads to duplication, delays, redundant stock, and avoidable downtime.

With real-time, network-wide coordination:

• Maintenance would align with planned upgrades
• Procurement would factor in actual multi-site inventory and manufacturer lead times
• New infrastructure would be designed to be compatible with existing components
• Inventory systems would dynamically reveal surplus stock and avoid unnecessary orders

The results include lower downtime, reduced costs, optimized resources, and higher resilience.

Data as a Natural Resource

Data should be treated as a natural resource, similar to how chemical elements combine to create new compounds with properties neither element possesses alone. When you pair the right datasets, you create insights and capabilities that are impossible with either dataset in isolation.

Even with the right tools, collaboration can stall due to misaligned incentives or entrenched mindsets. Teams often focus on protecting SLAs, regulators prioritize compliance, and legacy structures reward isolated performance. In industries like rail and energy, these challenges extend across organizations: different operators, infrastructure owners, and maintenance providers may have competing priorities, making coordinated action even harder.

Technology alone isn’t the solution. We need to create spaces and processes that encourage collective problem-solving across organizational boundaries. To do this, someone needs to own the table: a neutral or trusted facilitator who can convene the right stakeholders, align goals, and guide the conversation toward shared outcomes:

• Facilitate collaboration: Bring stakeholders from all relevant organizations together to align goals, share data, and uncover opportunities
• Prioritize opportunity: Identify where combining datasets across organizations creates disproportionate value
• Think end-to-end: Shift focus from individual metrics to collective outcomes across the entire value chain

Deliberate ownership of the table is critical. Without a convening force to bring the right parties together and facilitate discussion, collaboration often stalls before it begins. When the table is owned and the conversation guided, organizations can unlock hidden potential by combining their data and perspectives across both internal and external boundaries.

The Strategic Value of Data

Data is a strategic asset. Organizations that treat it as a passive byproduct will fall behind, while those that embrace early data discovery, collaboration, and end-to-end thinking across all stakeholders will:

• Reduce downtime and maintenance costs
• Optimize inventory and resource allocation
• Predict and prevent failures
• Build more resilient, efficient operations

But it’s not enough to have the right tools or data. We must change the way we think about innovation and prioritization. We must move from a mindset focused on individual metrics, departmental KPIs, or incremental improvements to one that puts collaboration and data at the center of every problem we tackle.

We have two choices: continue with siloed operations and reactive decision-making, or reimagine how we prioritize, plan, and innovate to maximize value across the entire network.

The tools exist. The data exists. The opportunity exists. There’s no need to wait. Collaboration can start today, and the rewards of thinking differently about innovation and data are too significant to postpone.