AI's Greatest Value Lies in Collaboration, Not Automation

Original

In reality, the vast differences between systems often render alignment difficult to achieve, leaving collaboration heavily dependent on manual effort and prohibitively expensive. However, AI's collaborative capabilities are beginning to break this impasse. The new models of collaboration it enables are not only transforming workflows but also reshaping the nature of value creation and competition in fragmented industries.

The economic implications of AI are frequently interpreted through the lens of reduced prediction or production costs. Yet a more critical, and often overlooked, dimension lies in the significant reduction of collaboration costs. Here, "collaboration" refers not to its literal interpretation, but to the continuous process of translating outputs from diverse teams and systems into mutually intelligible and interoperable information. Its costs are embodied in the labor required for data conversion, the effort expended on aligning outcomes, and the considerable time consumed in meetings, coordination, and rework. Effective collaboration ensures that disparate individuals, tools, and data can function cohesively toward a shared objective; in essence, it enables projects to progress steadily rather than becoming mired in endless cycles of adjustment and negotiation.

Artificial intelligence is rendering collaboration both inexpensive and ubiquitous. It achieves this through two primary mechanisms: extracting structured frameworks from unstructured information and utilizing that structured information to drive task execution. Complementary components that were previously unable to interoperate due to incompatible data formats, terminologies, or workflows can now be seamlessly integrated without imposing uniform standards. For the first time, collaboration can occur without the prerequisite of broad consensus, thereby extending its applicability to specialized domains once considered too costly or complex to coordinate.

Historically, platforms designed to integrate fragmented systems through collaboration have encountered a fundamental constraint: their reliance on highly structured data and standardized interfaces, which presupposes consensus on protocols and processes. In practice, the profound heterogeneity across systems often makes such alignment unattainable, perpetuating reliance on manual intervention and sustaining high costs. AI's capacity for collaboration is now challenging this limitation. The novel forms of coordination it introduces are not merely altering operational workflows; they are redefining the foundations of value creation and competitive advantage in increasingly fragmented sectors.

1. Collaboration without Consensus

Consider the construction industry as an illustrative case. A typical building project necessitates the collaboration of multiple specialized teams, including architects, structural engineers, and contractors. However, each team operates with its own professional tools and distinct priorities: architects focus on spatial aesthetics, structural engineers emphasize load-bearing safety, and contractors prioritize construction sequencing and progress management. In this context, effective coordination becomes critical to project success. Construction frequently deviates from original designs, and if teams fail to align their adaptations in a timely manner, the inevitable consequences include rework, delays, and quality defects.

Previous attempts to address this issue through top-down standardization, such as mandating a unified platform or data model for all participants, have often proven unsuccessful. At best, they achieved limited integration in certain areas while inadvertently creating misalignments in others. Moreover, specialized teams naturally resist abandoning the tools that define their professional expertise. Designs and construction plans generated by different tools remain largely incompatible, and this lack of alignment directly translates into high collaboration costs.

Artificial intelligence offers a fundamentally different approach. Rather than imposing uniform tools or standards, it accepts fragmentation as a given and extracts information from disparate sources, including BIM software, spreadsheets, site photographs, emails, and annotated PDFs. Each of these sources captures only a partial perspective of the project. By integrating them, AI constructs a unified panoramic view of the project, enabling every participant to track progress across multiple dimensions, including design, engineering, scheduling, and compliance, in real time.

With this unified view, project managers can coordinate fragmented processes while respecting the distinct workflows of each team. They can also make cross-disciplinary trade-off decisions directly, without convening lengthy coordination meetings. For example, if an architect relocates a stairwell, the AI layer can automatically detect the impact on structural beams and immediately notify the structural engineer. AI thus functions as an invisible coordination layer, continuously monitoring and reconciling changes across teams and synchronizing execution through real-time translation.

Many companies have already recognized this opportunity. Trunk Tools, for instance, aggregates information from tools such as Autodesk, used by architects and engineers to create designs and plans, and Procore, used by contractors to manage on-site work. By interpreting drawings, specifications, and schedules, it generates a structured and searchable project archive. This eliminates the need for teams to spend time cross-checking and correcting discrepancies; all queries can be directed to a single source of truth, which provides accurate answers based on the most current documentation. Industry leaders are also following suit. Procore's acquisition of DataGrid, for example, was driven precisely by its analogous capability to extract structured data frameworks from unstructured documents.

2. Reshaping the Competitive Landscape

When collaboration no longer depends on consensus, the moats of companies that have built their advantages on unified standards, interfaces, or processes begin to erode. The U.S. auto insurance claims market offers a telling example, where the startup Tractable is challenging the seemingly unassailable position of the industry giant CCC Intelligent Solutions.

Auto insurance claims involve multiple independent parties, including insurers, repair shops, parts suppliers, and appraisers, each using different systems and operating under distinct incentives. CCC has dominated the market by establishing a standardized set of damage codes and digital workflows, a common language widely adopted across the industry. To displace CCC, a competitor would need to overturn this standard, which would entail replacing existing tools, retraining personnel, and rewriting insurer processes. The associated switching costs are prohibitively high. For the past two decades, CCC's position appeared impregnable.

However, the competitive dynamics have shifted. In challenging CCC, Tractable has bypassed the battle over standards entirely. Rather than attempting to convince the industry to abandon CCC's coding system, it has trained AI models to directly interpret vehicle damage photographs taken by policyholders' mobile phones, generate repair estimates, and seamlessly integrate with insurers' existing workflows. In other words, Tractable has achieved collaboration without consensus. The results have been striking: by 2023, it was processing nearly 7 billion dollars in claims annually.

3. The Path Forward

The opportunities presented by AI-powered collaboration extend far beyond the industries discussed above. In fragmented sectors such as healthcare and logistics, artificial intelligence's capacity to reconstitute shared states across disconnected systems represents both an opportunity and a challenge. Incumbent firms have three strategic options to consider.

Becoming the Collaboration Layer

Firms pursuing this path acknowledge that proprietary standards alone are no longer sufficient to sustain competitive advantage. They will invest in constructing the most comprehensive panoramic view of the ecosystem. The logistics platform project44 exemplifies this strategy, offering unified visibility across freight shipments without requiring carriers to replace their existing standards. By embracing openness to diverse tools and formats, such firms position themselves as indispensable translation hubs within their ecosystems.

Strengthening End-to-End Accountability

Rather than competing for the role of a neutral collaboration platform, firms can instead reinforce their capacity to assume ultimate responsibility for outcomes, even when specific collaborative processes are executed externally or by third parties. Maersk exemplifies this strategy through its transformation into an integrated logistics service provider, offering end-to-end services encompassing contracts, customs clearance, warehousing, digital booking, and risk management. Amid increasing supply chain volatility, sanctions risks, and regulatory scrutiny, Maersk leverages its integrated control to deliver more reliable risk mitigation and execution assurance, thereby building its competitive edge.

Controlling Collaboration through Layered Access

Under this strategy, firms neither fully open their collaborative capabilities nor retreat entirely into closed control. Instead, they leverage AI to construct a privileged unified view internally, then selectively and conditionally, often for a fee, provide partial visibility to their still fragmented partner ecosystems.

FedEx pursues this approach. It has invested heavily in AI-powered routing systems, forming a unified operational view based on real-time data across its entire network. However, it shares information externally only to a limited extent, frequently behind paywalls. This compels customers and partners to rely on FedEx for critical data, thereby consolidating its hub status. As long as its network density and service coverage remain advantageous, customers will continue to pay for this visibility rather than building their own systems, sustaining this business model.

AI-driven collaboration, enabling people, tools, and systems to work together efficiently without consensus, will deliver tangible gains in speed, cost, and innovation for fragmented industries. It will also precipitate a shift in power. In the short term, the winners will be those who effectively deploy AI to bridge systemic divides and accomplish tangible work. Yet in the long run, collaboration alone will prove insufficient. As project scale expands and the costs of failure mount, concerns over accountability, trust, and ownership will inevitably drive industries to rebuild shared rules and clarify rights and responsibilities. A viable path forward involves using artificial intelligence to initiate collaboration where coordination costs have long been prohibitive, while consciously constructing governance mechanisms, contracts, and norms to transform rapid coordination into durable, trustworthy ecosystems.