Beyond the headlines lies a deeper, more consequential question: what does this acquisition reveal about the real prospects of AI in the AEC industries? Is AI poised to deliver transformative productivity gains, or is it simply the latest chapter in a long history of digital ambition colliding with structural reality?

To answer that question, it is not enough to assess Consigli AS technology or AECOM’s balance sheet. We must first confront a more uncomfortable truth: digital transformation in AEC has always struggled—not just because of insufficient ambition or technical skill, but also because of the fundamental way the industry operates.

Why Digital Transformation Keeps Stalling in AEC

The AEC industry consistently ranks near the bottom of cross-sector productivity and digital-impact benchmarks, often alongside agriculture. This is surprising for a sector that designs and delivers trillions of dollars’ worth of assets each year and employs some of the world’s most advanced technical talent.

The usual explanations—lack of standardization, limited modularization, cultural conservatism—are not wrong, but they are incomplete. The deeper issue is structural variability, baked in at every level.

Every project begins with a unique physical context. Soil conditions, climate, hydrology, topography, and local material availability fundamentally shape what can be designed and built. A foundation solution that works in the Netherlands is inappropriate in California; timber may be optimal in Scandinavia and impractical elsewhere. These are not inefficiencies—they are rational responses to physical reality.

Layered onto this is regulatory diversity. Building codes, safety standards, planning regimes, procurement models, and permitting processes vary not just by country, but often by region or municipality. Each reflects local history, risk tolerance, and governance structures. The result is not one AEC market, but thousands of overlapping regulatory micro-contexts.

Finally, organizational fragmentation reinforces variability. Responsibility transitions repeatedly across the asset lifecycle—from planning to design to construction to operations—often across different firms with different incentives, tools, and data structures. Project teams are assembled anew for each assignment, dominated by SMEs, and disbanded once delivery is complete. Continuity is the exception, not the rule.

These conditions are not pathologies; they are the natural consequence of building complex, bespoke assets in diverse environments. But they create a profound challenge for digital technologies that depend on scale, repeatability, and uniformity.

The Scaling Problem: Lessons from Past Failures

The AEC industry has seen repeated attempts to “industrialize” construction through technology. Two of the most prominent—Katerra and WeWork—illustrate why scale is so elusive.

Katerra, backed by more than $2 billion from SoftBank, attempted to vertically integrate design, manufacturing, and construction through highly automated, standardized workflows. Despite enormous capital and technical ambition, it collapsed in 2021. Factories underperformed due to inconsistent demand, cost overruns mounted, and project variability proved resistant to standardization.

WeWork’s “Powered by We” initiative followed a similar trajectory. By standardizing interior fit-outs and leveraging BIM to accelerate global rollouts, the company sought to productize construction at scale. But site conditions, building constraints, and tenant variability undermined repeatability. High design costs and operational complexity overwhelmed the promised efficiencies.

The pattern is familiar. The technology is feasible. The desire is strong. But the business model fails the viability test because the user base is too small or too inconsistent to support a standardized product.

This is why comparisons to manufacturing are misleading. Manufacturing operates in controlled environments producing standardized products at scale. AEC delivers unique assets under variable conditions. Expecting identical digital dynamics is like comparing apples to oranges.

Where Digital Has Worked—and Why

Despite these challenges, digital technology implementation has not failed in AEC. It has succeeded precisely where flexibility, not standardization, is the organizing principle.

Major platform vendors—Autodesk, Bentley, Nemetschek, Trimble, ESRI—have thrived by offering tools that can accommodate almost any project condition, regulation, or workflow. Their success is not due to rigid best practices, but to configurable frameworks that allow users to adapt technology to context.

Similarly, some of the most effective digital innovation happens quietly at the project level. Portfolio-based automation, citizen development using low-code tools, and narrowly scoped AI applications have delivered real value—even if they rarely scale across the enterprise.

At Arcadis, for example, AI-based image recognition was initially developed to automate wildlife monitoring on infrastructure projects. It later expanded into transportation asset management and eventually became a mature digital product used beyond the AEC sector. The lesson is not that every in-house tool should scale, but that learning, differentiation, and capability-building have value—even when products are eventually retired.

Enter Generative AI—and a Familiar Hype Curve

The current wave of AI enthusiasm feels different because, in many ways, it is. Generative AI has crossed a threshold from academic promise to practical utility. Tools like ChatGPT have demonstrated human-like interaction at scale, triggering massive investment and accelerating adoption across industries.

Yet the broader dynamics remain familiar. AI appears to be sitting near the peak of the classic hype curve: extraordinary expectations, enormous capital inflows, and limited evidence—so far—of sustained, industry-wide returns.

For AEC, AI adoption is already following historical patterns. Flexible, general-purpose tools are spreading rapidly. Highly specific automations show promise but struggle to scale. The constraint is not technological capability, but economic viability in a fragmented, project-based market.

There is also a fundamental technical limitation that receives little attention: most AI models are trained on raster data (images, text), not the vector-based formats that underpin CAD, BIM, and GIS. AI can assist with scripts, documentation, and optimization, but it cannot yet natively generate or reason over complex, editable vector models. For now, AI remains an assistant, not a replacement, for core design platforms.

The Consigli AS Acquisition: Strategic Logic Meets Structural Reality

Against this backdrop, AECOM’s acquisition of Consigli AS makes more sense but also looks more risky.

Consigli AS positions itself as “The Autonomous Engineer,” promising dramatic reductions in engineering time through AI-driven automation of layouts, calculations, BIM models, and documentation. If even a fraction of these claims hold at scale, the competitive advantage would be significant.

AECOM’s stated rationale is clear: proprietary technology, strategic differentiation, accelerated innovation, and access to elite AI talent. In the short term, these benefits are real. The acquisition gives AECOM visibility, credibility, and a head start in learning how AI reshapes engineering delivery.

But history suggests that sustaining this advantage will be difficult.

A proprietary, in-house AI platform faces several structural headwinds. The internal user base—even at AECOM’s scale—may be too small to justify long-term platform investment. Competing vendors are embedding similar capabilities into off-the-shelf tools, eroding differentiation. Data readiness remains uneven. And the economics of time-and-materials contracts limit how much efficiency can be monetized.

Market reaction reflects these concerns. The immediate drop in AECOM’s share price suggests investor skepticism about near-term ROI. Reports that some Consigli AS customers exited following the acquisition underscore persistent anxieties about data ownership and competitive exposure.

What This Means for the Industry

AECOM’s move should not be dismissed as reckless. It is a calculated bet—one that prioritizes learning, positioning, and early-mover advantage in an uncertain landscape. But it also exposes a broader truth: AI will not magically resolve the structural constraints that have limited digital transformation in AEC for decades.

Real value will come not from platform ownership alone, but from organizational alignment, data maturity, and the ability to identify and scale genuinely repeatable use cases. The winners will be those who treat AI not as a silver bullet, but as a flexible capability embedded within adaptable operating models.

For most AEC firms, this points toward a pragmatic strategy: invest in people, data foundations, and change management; leverage established platforms rather than building everything in-house; and focus differentiation on domain expertise and client value—not proprietary technology for its own sake.

A Bold Signal, Not a Final Answer

AECOM’s acquisition of Consigli AS is best understood as a signal rather than a conclusion. It signals that AI is now strategically unavoidable. It signals that scale and learning matter more than perfection. It signals that the next phase of digital transformation in AEC will be shaped as much by economics and organizational design as by algorithms.

Whether this particular bet pays off remains uncertain, but the debate it has sparked—about realism, risk, and the true nature of innovation in AEC—is one construction and the professional services that provide its designs can no longer afford to avoid.

Brian Mommers is an independent strategic advisor and non-executive director focused on governance, digital and AI transformation, and leadership. As Global Technology Officer at Arcadis, he led enterprise-wide digital change for 36,000 employees, launching AI, federated data governance, and Agile teams. He brings a collaborative style and a passion for helping organizations navigate change with clarity and confidence.

Alison Jones is a business and technology executive with more than 30 years of experience in the Architecture and Engineering industry. She has led operations, digital transformation, and global technology initiatives. Her focus is on aligning technology investment with business outcomes and client value. Formerly with Arcadis, Alison is currently the CEO of Order Penguin, an AI powered equipment rental platform.