Hitachi Digital Services – an integrated digital powerhouse is awakening

PAC shares key observations from an exclusive briefing day in Lisbon, Portugal, exploring the convergence of OT, IT, AI, and engineering capabilities.

Jennifer Jonat, Arnold Vogt, and Wolfgang Schwab from PAC were invited by Hitachi to an exclusive briefing in Lisbon, Portugal. The discussion was not a standard corporate update. It provided a broad, detailed overview of how Hitachi Digital Services and GlobalLogic are positioning themselves at the intersection of enterprise IT, operational technology, product engineering, artificial intelligence, sustainability, and industrial transformation.

The briefing’s strongest message was clear: Hitachi Digital Services is becoming an integrated digital giant with exceptional engineering capabilities and a strong focus on AI. The company is building its narrative around AI as an execution capability, embedded across products, factories, assets, operations, sustainability processes, and enterprise platforms. This is an important distinction. Many providers still talk about AI mainly in terms of productivity tools, copilots, or experimentation. Hitachi Digital Services has the ambition to embed AI into operational environments where downtime, quality, compliance, safety, and energy use have direct business impacts. To further expand its AI capabilities, Hitachi acquired Synvert in 2025, an AI consulting firm headquartered in Germany with 550 employees, and integrated it into its GlobalLogic subsidiary. Proof that the integration is going well lies in the fact that the Synvert team already sits on the same floor as the broader Hitachi team in Portugal and actively participated in the briefing session.

The scale argument behind Hitachi’s positioning

Hitachi’s digital positioning is built on a combination of group scale and operational heritage. The company highlights 269,000 employees, a presence in more than 70 countries, over 300 engineering centers, more than 10,000 customers, 85% of the Fortune Global 100 served, and 182,000 global patents. It also frames its business around three large group domains: Digital, Green Energy & Mobility, and Digital Connective Industries.

This matters for two reasons. First, Hitachi’s credibility in AI-enabled industrial transformation depends less on generic AI skills and more on its ability to integrate engineering, IT delivery, operational knowledge, and repeatable platforms. The company emphasizes its 110+ years of operational technology experience, 60+ years of IT experience, 20+ years of AI transformation, and 25+ years of digital product engineering. Its stated operating logic is to help companies build, modernize, and scale, with a focus on intelligent products, intelligent enterprise, and intelligent operations. Second, Hitachi, with its more than 200 factories, often acts as “customer zero” for new Hitachi Digital Services’ services. This is a clear advantage in the market that strongly supports the development and testing of new digital solutions in real-world industrial scenarios.

This framing makes sense. Most large industrial organizations are not seeking another AI proof of concept. They are working to modernize legacy estates, integrate fragmented IT and OT landscapes, improve asset performance, meet regulatory requirements, and address workforce constraints. AI becomes relevant only when it is tied to these issues.

Build, modernize, scale: a useful but demanding structure

Hitachi Digital Services’ “build, modernize, and scale” structure gives the portfolio a clear logic:

“Build” covers digital product design, embedded and connected systems, data and AI engineering, and cloud-native platform engineering.
“Modernize” focuses on enterprise platforms, IT-OT-AI integration, data migration, cloud modernization, and reducing legacy complexity.
“Scale” covers IT operations, AI-driven site reliability engineering, AI and machine learning operations, DevSecOps, FinOps, and RunOps.

The advantage of this structure is that it avoids treating AI as a separate consulting topic. Instead, AI becomes part of engineering, modernization, and operations. This is where enterprise demand is headed. Buyers increasingly ask how AI changes software engineering, platform operations, service management, quality control, maintenance, energy optimization, and compliance reporting. Hitachi seems to understand this shift.

The challenge will be to achieve commercial clarity. A broad portfolio can be hard to buy if it is not packaged into specific entry points. The most promising entry points are likely to be those where the business case is tangible: faster software engineering, lower run costs, fewer incidents, faster cloud migration, better energy visibility, reduced manual ESG work, predictive maintenance, quality inspection, and industrial asset monitoring.

AI as an engineering discipline, not just an interface

The AI section of the briefing was most convincing when it moved beyond generic copilots and toward the engineering discipline. Hitachi’s AI approach combines trustworthy AI at scale through its R2O2.ai framework, production reliability for AI operations through HARC for AI, and software engineering acceleration through VelocityAI. Together, these capabilities support AI-powered software delivery, agentic workflows, and Physical AI. The focus is not only on building models or using large language models, but also on securing, governing, operating, and improving AI systems in production.

This is important because enterprise AI is entering a more mature phase. The first wave was dominated by experimentation and user-facing assistants. The next phase will focus on operating AI safely in production environments. Hitachi’s AI observability view covers infrastructure metrics, model performance, data reliability, safety metrics, functional metrics, cost, and even carbon footprint. That is the right direction. AI systems will need the same operational discipline that enterprises already expect from cloud platforms, applications, and infrastructure, plus additional controls for model behavior, data drift, prompt injection, toxicity, privacy leakage, and explainability.

The discussion of agentic AI was also more balanced than much of the current market messaging. Hitachi and GlobalLogic distinguish between deterministic workflows and agentic architectures. Deterministic workflows are generally better suited to predictable, high-volume processes where consistency, compliance, and control are important. Agentic AI becomes more valuable when work is variable, context-heavy, poorly documented, or requires reasoning and orchestration across multiple systems. This distinction is essential. Not every workflow should become agentic, and not every decision should be delegated to autonomous agents.

Agentic AI needs governance before scale

GlobalLogic’s agentic AI view centers on a shift from linear workflows to autonomous value loops. Agents are expected to interpret goals, evaluate context, use tools, collaborate with other agents, and adapt workflows dynamically. This shift creates opportunities in software engineering, cloud and data migration, autonomous DevOps, intelligent FinOps, security operations, industrial processes, supply chain optimization, and managed services. Hitachi also references a library of more than 200 prebuilt agents, an agent management system, and a control layer for governing, securing, and observing agents across platforms.

However, the briefing also made clear that this cannot scale without governance. Agentic AI increases the need for accountability, transparency, fairness, security, privacy, and compliance. The security discussion covered prompt injection, data poisoning, model inversion, adversarial attacks, unauthorized actions, and supply chain vulnerabilities. The proposed principles, including Zero Trust AI, least privilege, explainability, continuous monitoring, and secure multi-tenant environments, are not optional; they are prerequisites for agentic systems to operate in regulated industries or OT-adjacent environments.

From PAC’s perspective, this is one of the more relevant parts of Hitachi’s story. AI governance is often discussed as a policy topic. Hitachi is trying to connect it to architecture, runtime control, observability, and managed operations. That is where governance becomes operational.

Physical AI and Industry 5.0: where Hitachi’s heritage matters

The most distinctive part of the Lisbon briefing was the discussion of Industry 5.0 and Physical AI. Hitachi uses the term to describe the convergence of IT, OT, and AI in environments where digital intelligence interacts with physical assets, production systems, infrastructure, and workers.

The logic is straightforward: industrial organizations need to understand what should happen, what is happening, what can be sensed, what needs to be done, and how execution can be automated or assisted. Hitachi integrates digital product lifecycle management, OT systems, asset lifecycle management, vision intelligence, telemetry, edge AI, agentic AI, and closed-loop automation with manufacturing operations management, OT, and edge platforms. The target outcomes are servitization, operational efficiency, and reliability.

The use cases discussed were concrete: AI-supported surface defect detection, root cause analysis, shop-floor assistance, robotic process control, intelligent production scheduling, self-optimizing industrial processes, maintenance bots, digital thread and PLM integration, supply chain optimization, virtual factories, synthetic data generation, and digital twin-based process optimization. The industry focus is particularly relevant to manufacturing, energy and utilities, mobility, and rail.

This is where Hitachi’s positioning differs from that of many pure IT service providers. Physical AI requires an understanding of machines, field environments, safety constraints, industrial data, engineering workflows, and real-time operations. It also requires integration with existing MES, MOM, PLM, SCADA, edge, cloud, and enterprise systems. This is difficult territory, but it is also where generic AI providers are weakest.

NVIDIA, AI factories, and industrial compute

Hitachi’s partnership with NVIDIA is positioned as a key accelerator for industrial AI. The briefing included examples such as AI factories, industrial digital twins, autonomous tram development, predictive maintenance, real-time decision support, edge AI computing, synthetic data, and GPU-based workloads. Hitachi also uses the AI factory concept as a secure, scalable, and centralized environment for AI infrastructure, with attention to GPU utilization, demand and supply planning, forecast accuracy, and cost per GPU.

The strategic relevance is clear: industrial AI will need more than models; it will require compute infrastructure, simulation environments, edge deployment, model lifecycle management, data pipelines, and domain-specific validation. The AI factory concept could offer a practical solution to one of the main constraints in industrial AI: how to move from individual pilots to repeatable, governed, and scalable development.

Sustainability as data, operations, and compliance

The sustainability discussion broadened the overall picture. Hitachi is not only discussing sustainability reporting. Its portfolio spans product lifecycle intelligence, smart operations and decarbonization, ESG data and regulatory intelligence, sustainable digital infrastructure, sustainable supply chains, and workplace safety and risk prevention.

This is a sensible portfolio because sustainability has become an operational data challenge. Companies need lifecycle assessments, product carbon footprints, environmental product declarations, digital product passports, Scope 3 visibility, supply chain traceability, energy management, data center carbon visibility, and regulatory reporting. These needs are increasingly interconnected. A company cannot credibly report its sustainability performance if product, supplier, operational, IT energy, and regulatory data remains fragmented.

The RitaONE platform exemplifies this direction. It is positioned as an ESG intelligence platform offering automated data collection, framework interoperability, emissions calculation, audit-grade traceability, AI-generated insights, and natural language access to ESG data. Reported capabilities include more than 6,200 predefined ESG KPIs, 60% less time spent on data collection, aggregation, and internal reviews, and three times faster creation of compliant ESG reports.

The Hitachi Rail example was particularly useful because it showed what scale can look like: 214 entities, 100 facilities, 114 legal entities, 109 ESG topics, more than 6,000 KPIs, 204 stakeholders, and use across 58 countries. The reported outcome was about 60% less manual effort and three times faster reporting cycles.

Green IT and carbon visibility below the surface

Another relevant topic was sustainable digital infrastructure. Many companies still estimate IT emissions using rough proxies. Hitachi’s approach aims to go deeper, with real-time monitoring across data centers, networks, and cloud environments. The Green IT model covers power supply and equipment, IT infrastructure, applications and services, architecture and engineering, cloud strategy, and data stewardship.

The key point is granularity. If companies can attribute energy consumption and carbon emissions to processes, applications, services, or workloads, they can make better decisions about optimization, modernization, workload placement, procurement, and architecture. This is no longer only a sustainability issue. It affects IT costs, infrastructure strategy, application streamlining, and cloud governance.

The analyst view: strong ingredients, now the packaging matters

The Lisbon briefing confirmed that Hitachi has strong ingredients for a differentiated market position. The combination of industrial heritage, engineering depth, GlobalLogic’s software and product engineering capabilities, AI governance, sustainability platforms, OT know-how, and operational use cases is credible. The company is strongest where digital technology intersects with physical operations: factories, rail, energy infrastructure, mobility, industrial assets, and complex supply chains.

The open question is how convincingly Hitachi can package this for enterprise buyers. The integrated IT-OT-AI story is strategically compelling, but it must be easy to understand, easy to buy, and easy to scale. Buyers will not purchase “Industry 5.0” as an abstract concept. They will buy lower downtime, faster production cycles, better quality, lower energy costs, safer operations, compliant reporting, faster engineering, and more reliable AI operations.

For PAC, the key takeaway is that Hitachi should not be viewed as just another digital services provider adding AI to its portfolio. Its strongest proposition is more specific: AI-enabled industrial and operational transformation, grounded in engineering, OT, enterprise platforms, and sustainability data. This is a narrower story, but it is also a stronger one.

To explore more insights into Hitachi Digital Services’ strategy, read our blog posts on: