作者:Steve McDowell
Arrow McLaren IndyCar sponsored by NTT DATA.
Steve McDowell
The pit area of an IndyCar race is an unlikely place to learn about the challenges of enterprise AI. Yet, it was in a trailer housing the Arrow McLaren IndyCar team at the recent Java House Grand Prix near Monterey, California, where I did just that.
Calling it a trailer is a disservice. This two-story rolling headquarters houses over a dozen engineers and data scientists crunching data in real-time as the team’s driver, Silicon Valley native Nolan Siegel, takes laps on the track just outside in NTT Data-sponsored car number 6.
IndyCar racing is the perfect metaphor for, and ideal demonstration of, the power and promise of AI. There are very few enterprise applications, after all, that move at the velocity of a 180-mile-per-hour race car, streaming billions of data points each hour from over 150 sensors.
Despite all that data, as chief engineer of car number 6, Kate Gundlach, told me, “The driver is the ultimate sensor.” This isn’t dissimilar to business, where, as Ms. Gundlach went on to say, "Having no data is better than having bad data."
The enterprise AI revolution is at an inflection point. As generative AI transitions from experimentation to real-world deployment, agentic AI is the next frontier. Yet for most enterprises, the path from AI ambition to execution remains fraught with complexity. A recent MIT report revealed that 95% of generative AI pilots have, thus far, failed to produce meaningful results.
The challenge isn't just technological. Large language models, hyperscaler cloud platforms, multimodal data integration, and security frameworks rarely work together seamlessly.
Enterprises need to orchestrate these tools across legacy systems while ensuring regulatory compliance, managing data sovereignty, and delivering measurable ROI. Few organizations possess the expertise or resources to tackle this transformation alone.
This is where global system integrators come into play. GSIs, long known for steering digital transformations, are becoming the connective tissue that makes enterprise AI operationalizable. And among the leaders charting this path, Tokyo-based NTT Data is making some of the most strategic moves in the market.
Enterprise AI adoption differs fundamentally from previous technology waves in two critical ways. First, it requires full-stack transformations that span data pipelines, cloud infrastructure, security, applications, and business workflows. Second, it requires ecosystem orchestration, as no single vendor can address every enterprise need.
Andrew Wells, chief data and AI officer for NTT Data, told me that GSIs bridge these gaps through three core differentiators:
In essence, GSIs function as AI adoption accelerators, translating innovation into measurable enterprise transformation.
NTT Data, with operations spanning over 50 countries, exemplifies how GSIs can capitalize on the AI moment. Over the past year, the company has announced major partnerships with Google Cloud, Microsoft, Mistral AI, Salesforce, and Corvic AI.
Gartner recently recognized NTT Data's momentum, naming the company an "Emerging Leader" in its 2025 Innovation Guide for Generative AI Consulting and Implementation Services. Gartner said that this reflects NTT Data’s blend of innovation, market understanding, and delivery capability.
Sudhir Chaturvedi, CEO of North America and global chief growth officer for NTT Data, explained that this is all by design. The GSI has transitioned from a service market participant to a shaper of the enterprise AI landscape. This only happens, he said, “if we can deliver the full stack” and become “part of each client’s fabric.”
Collectively, these are moves that position NTT DATA as one of the most aggressive integrators in the generative and agentic AI ecosystem.
NTT Data recently entered into a global partnership with Google Cloud to accelerate AI-powered cloud innovation. The collaboration builds on their 2024 APAC co-innovation agreement and NTT’s acquisition in late 2024 of Google Cloud specialist Niveus Solutions.
The global collaboration between the two companies focuses on the adoption of agentic AI and cloud-native modernization, addressing two of enterprises' most pressing challenges: regulatory compliance and operational scalability.
The partnership delivers concrete industry solutions. In the financial services industry, NTT Data’s Regla platform operates on Google Cloud to streamline regulatory reporting and compliance. In hospitality, its Virtual Travel Concierge leverages Google's Gemini models for real-time, multilingual customer service, handling over three million conversations per month.
Sovereign AI capabilities are another strategic pillar. Using Google Distributed Cloud, NTT Data helps clients deploy air-gapped and connected sovereign cloud environments, a critical capability for highly regulated industries navigating data residency requirements.
To support the collaboration, NTT Data created a dedicated Google Cloud Business Group staffed by “thousands” of engineers and architects, with plans to certify 5,000 additional professionals in Google Cloud technologies.
This alliance reveals a lesser-known, broader industry trend: hyperscalers require system integrators as much as integrators require hyperscalers. Microsoft, which has recently entered into a new partnership with NTT Data, also collaborates with Accenture and Avanade. Similarly, Amazon Web Services partners with Deloitte.
Google Cloud's deepened partnership with NTT Data signals that competition extends beyond workloads, instead influencing how enterprises structure their AI transformations.
While many enterprises remain in generative AI pilot mode, NTT Data is already addressing the next frontier of multi-agent AI systems. This involves multiple specialized AI agents collaborating on complex tasks, such as IT service management or personalized insurance support.
Through its Microsoft partnership, NTT Data has developed managed services that orchestrate multi-agent workflows on Azure AI Foundry. At the recent Microsoft Build conference, the company demonstrated a multi-agent ticket management system that reduced response times and improved customer satisfaction through automated classification, prioritization, routing, and resolution.
The business case is compelling: streamlined workflows, enhanced compliance, and significant cost savings through intelligent automation. As enterprises transition from experimentation to scale, orchestration frameworks such as this become critical enablers of adoption.
NTT Data is also hedging against hyperscaler dependence through its partnership with Mistral AI, the Paris-based startup behind high-performance, open-weight generative models. Together, the two companies are co-developing secure, private AI platforms for regulated sectors including finance, insurance, and defense.
Early implementations include a sovereign AI platform in Luxembourg and a patent-search application for Dennemeyer, a global IP services firm. By offering private, sustainable AI deployments, the NTT Data-Mistral alliance addresses the rising demand for strategic AI autonomy.
Additional partnerships further strengthen NTT Data’s position. New services for Salesforce's Agentforce platform enable enterprises to augment their teams with autonomous AI agents, leveraging NTT Data’s "Evangelize, Pilot, Adopt, Scale" methodology.
Meanwhile, an alliance with Corvic AI extends capabilities into multimodal data management. Corvic's platform reconciles data from documents, graphs, tables, and images to solve the problem of fragmented enterprise data, a common bottleneck in AI adoption.
The enterprise AI market is entering its scaling phase. IDC projects global AI systems spending will surpass $500 billion by 2027. Yet, surveys consistently show that enterprises struggle to move beyond the proof-of-concept stage, facing familiar roadblocks such as talent shortages, integration challenges, security concerns, and uncertain ROI.
GSIs, such as NTT Data, are emerging as the solution to this execution gap. While companies like OpenAI, Meta, and the hyperscalers lead the way in building models, it's GSIs who help enterprises ensure those models deliver a measurable business impact across industries.
Agentic AI strengthens this case. Orchestrating multiple AI agents across vendors and platforms requires the systems expertise, governance frameworks, and global delivery models that GSIs have refined over decades.
The next 18 months are pivotal. AI model builders will continue to release new foundation models, while open-weight players like Mistral will increasingly push alternatives. Enterprises face mounting pressure to demonstrate AI ROI at scale, even as regulators tighten oversight around data privacy, security, and sovereignty.
In this environment, GSIs will become essential partners. NTT Data’s recent partnership momentum and client wins demonstrate what's possible when GSIs align technical expertise, ecosystem breadth, and industry-specific delivery capabilities.
Sitting trackside at the WeatherTech Raceway Laguna Seca, Arrow McLaren team principal Tony Kanaan summed it up, saying, "Once, the driver taught the computer; now it’s the opposite." In working with NTT Data, he said, "we're all learning together" – words that could be spoken about nearly any modern enterprise AI effort.
Disclosure: Steve McDowell is an industry analyst, and NAND Research is an industry analyst firm, that engages in, or has engaged in, research, analysis and advisory services with many technology companies, but has no business relationship or financial interest with any company mentioned in this article. No company mentioned was involved in the writing of this article.