AI-native disruption is drawing nearer, but Nemetschek still retains short-term system-of-record defensibility

Overview

This report discusses the potential disruption paths of AI in AEC software. Bernstein believes that AI-native platforms such as Claude Design and Prometheus are changing user expectations for the interface, speed, and delivery model of design and engineering software, but these platforms are not currently direct substitutes for Nemetschek's core BIM and system-of-record products. Therefore, the short-term investment conclusion remains neutral, while the long-term strategic risk is becoming clearer.

Core views

The core views are: first, AI-native competition is more concentrated in adjacent layers, so the short-term impact is indirect; second, industry value may shift from tool-based software licensing to outcome-based delivery, meaning value capture could decline even if traditional software continues to be used; third, the key data moat has shifted from static project files to real-time feedback loops embedded in workflows; fourth, Nemetschek's AI direction is reasonable, but at present it looks more like a gradual defensive move and has not yet proven it can translate into measurable product differentiation or pricing power.

Analysis framework

The report uses a two-dimensional framework of 'traditional vendors vs. AI-native players' and 'tools vs. outcomes' to analyze how competitive pressure does not need to appear as direct substitution, but can also erode the valuation and value capture of traditional software vendors through budget reallocation, user interface migration, and outcome-based services.

Methodology notes

• Competitive landscapeTraditional vendors vs. AI-native players; tools vs. outcomes

  Two-dimensional competitive framework

  This framework places Nemetschek in the quadrant of traditional vendors plus tools, while AI-native value creation flows more toward the quadrant centered on outcome delivery, and is used to assess indirect but structural competitive pressure.

• Data moatStatic data vs. live data and feedback loops

  Sustainable AI data advantage

  The report argues that merely owning large volumes of project files no longer constitutes a strong moat; true defensibility comes from a closed-loop system that continuously records user decisions, feedback, and downstream outcomes.

• Scenario analysisBull and bear cases

  Balancing short-term resilience and long-term disruption

  The bull case emphasizes system-of-record stickiness, regulatory responsibility, and human-machine collaboration; the bear case emphasizes AI-native engineering autopilot, outcome-based services, and commoditization pressure.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Nemetschek (NEM.GR)

  Core covered company, positioned in traditional AEC software and BIM systems of record.

  Strengths

  Deeply embedded in AEC workflows, with defensibility from system-of-record stickiness, a portfolio of brands, and the complexity of regulation and collaboration.

  Weaknesses

  The path to AI monetization remains unclear; current AI capabilities are more assistive and incremental, and the closed-loop data moat has not yet been fully proven.

  Comparison

  Compared with AI-native platforms, Nemetschek is stronger in its existing customers, workflows, and data entry points; compared with larger incumbent vendors such as Autodesk and Bentley, it also faces pressure from accelerating AI investment and feature commoditization.

  Risks

  Outcome-based AI services eroding budgets, commoditization of the tool layer, reasoning and integration costs compressing margins, and user interfaces being redefined by AI-native layers.

• Claude Design

  An AI-native design and prototyping platform example, representing the direction of upstream design exploration and multimodal interaction.

  Strengths

  Prompt-driven, multimodal interaction, rapid generation and iteration, potentially changing the benchmark for user experience in design tools.

  Weaknesses

  It is not currently specifically aimed at core AEC systems of record, and evidence of direct short-term substitution for Nemetschek is limited.

  Comparison

  Compared with Nemetschek's BIM and traditional toolchain, it emphasizes reconstruction of interfaces and creative paradigms rather than existing files, collaboration, and regulatory processes.

  Risks

  If user habits shift to AI-native interfaces, the user stickiness and pricing power of traditional tools may be weakened.

• Prometheus

  An AI-native engineering automation example, representing the direction of outcomes and engineering autopilot.

  Strengths

  Strong capital backing, aiming to automate the design and prototyping of complex physical systems, and potentially shifting engineering work from tool usage to outcome delivery.

  Weaknesses

  Its current focus is on industrial and physical product engineering rather than BIM-centered architectural design, and entering AEC still requires overcoming standards, localization, and regulatory barriers.

  Comparison

  Unlike Nemetschek, which sells software tools, the model represented by Prometheus is more likely to capture budget by delivering faster, cheaper, or higher-quality engineering outcomes.

  Risks

  If similar capabilities expand into AEC, high-value engineering judgment could be absorbed by the AI outcome layer, with traditional software degrading into validation, packaging, or compatibility infrastructure.

Key data

• RatingMarket-PerformThe report maintains a neutral rating, believing the risk-reward is broadly balanced.
• Target price104.00 EURPrice Target disclosed on the cover page.
• Prometheus funding and valuationApproximately $12 billion in funding and an approximately $41 billion valuationAn example of accelerating capital investment in AI-native engineering automation platforms.
• Nemetschek AI AssistantInitially deployed in Graphisoft Archicad and Allplan in 2025, and expanded to VectorworksReflects the company's shift from standalone AI features to a unified AI foundation and assistant layer.
• Core investment conclusionIndirect short-term impact, with credible but gradual long-term structural riskSupports a Market-Perform rating rather than a directional rating change.

Impact & implications

The investment implication is that Nemetschek still benefits from system-of-record stickiness, workflow embedding, and regulatory barriers, but its valuation needs to reflect the medium- to long-term value migration risk posed by AI-native players. If the company cannot convert workflow data into closed-loop learning and outcome-based services, its products may become a low-value tool layer within broader AI workflows.

Risks

• AI-native outcome-based services reallocating customer budgets without directly replacing BIM.
• Static project files and generic AI features being commoditized by LLMs and third-party tools.
• Nemetschek has not yet proven that its closed-loop learning system can deliver sustained model improvement, product differentiation, or new revenue.
• AI integration, data management, and inference costs may make AI a defensive investment in the short term and compress margins.
• Large incumbent vendors such as Autodesk and Bentley, as well as AI-native players, are all accelerating investment, increasing competitive intensity.
• If AI-native orchestration layers control the user entry point, traditional AEC systems may be reduced to back-end infrastructure for data, formats, and compatibility.

What to watch

• Whether Nemetschek AI Assistant evolves from an assistive feature into automated design, compliance, or optimization outcomes.
• Whether the company discloses measurable evidence of closed-loop learning, customer feedback data usage, and model performance improvement.
• Whether AI features can bring standalone pricing, higher renewal rates, or product suite upgrades, rather than merely retaining customers.
• Whether AI-native platforms move into core AEC use cases such as BIM, structural design, building systems, and energy optimization.
• Whether customer budgets shift from software seats to outcome-based contracts, software-as-a-service, or automated engineering delivery.
• The impact of AI investment on gross margin, R&D expense, and inference costs.