From ambition to execution: What enterprise leaders are getting wrong about AI readiness
The AI race is on. Trillions of dollars in value are forecast to be created as AI disrupts industries, accelerates product development and reshapes decision-making. Analysts at Morgan Stanley predict that full AI adoption across the S&P500 alone could translate in time to $13-16tn in additional market capitalization.
But in a recent survey of 500 senior executives at large enterprises across North America, Europe and Asia, we found one common factor that’s stopping IT leaders from reaping AI’s rewards.
Put simply, executives are focused on AI applications and models, while paying less attention to the foundations that determine how much value can actually be captured.
“It’s not about just AI being an application or a tool. And a lot of people think about it that way,” says Sham Chatterjee, Former CIO. “The network matters because moving that data securely across your users, models, data centers and cloud platforms is what makes it work.”
Without getting these foundations right, scale remains out of reach and promising AI use cases struggle to become lasting advantage.
Five loops that determine AI success
Our research found that AI readiness depends on five interconnected dimensions: network infrastructure, systems integration, skills readiness, governance capabilities, and capital allocation. Together they determine whether AI investment compounds in value or plateaus over time.
Examining these loops highlights where organizations are falling short and what leaders need to do next.
Foundation loop
Start with the foundations. Sixty-five per cent of businesses we surveyed still operate legacy systems not designed for the data intensity and integration demands of enterprise AI. As a result, they often lack the networking, cloud and orchestration capabilities needed to scale reliably.
“You can run an agent in the data center and everything works,” says Rajarshi Purkayastha, Vice President at Tata Communications. “But when you take the agent out and give it to five users, 100 users, 10,000 users or a million users, it needs to behave in the same manner. It won’t work like that unless the infrastructure can support massive connectivity, low latency and minimal packet loss.”
That challenge is already becoming apparent for many organizations. As Seth Goodman, CRO at Boost Payment Solutions, explains: “Everyone I’m speaking to has an AI initiative. But if their systems are not AI-ready, then you just don’t see the productivity output.”
This is why organizations can experiment on outdated systems, but they can’t scale on them.
Integration loop
AI tools must integrate with legacy applications, cloud environments, and workflows. Yet many organizations still operate fragmented platforms and siloed data, challenges that often remain hidden during small-scale pilots.
Enterprises have to avoid scaling difficulties by connecting their systems now, through common standards, shared policies and clearer control layers. Modern infrastructure enables this sort of interoperability, and interoperability is what enables enterprise deployment.
Skills loop
Nearly one-third of leaders believe that skill gaps and a shortage of specialized talent is a primary roadblock to realizing value from AI initiatives. AI expertise too often remains concentrated within small technical teams in enterprises. But when skills are narrow like this, deployment and gains remain incremental.
“It’s not all about the data scientists,” continues Chatterjee. “If you’re implementing AI for the business, you need business owner participation, data engineers, cybersecurity, network engineers and architects. Legal is also your best friend. It takes a village to make all of this work.”
Instead, IT leaders must enable practical learning, role-specific guidance and support that
helps people use new tools inside real workflows. The goal is not to make everyone an AI expert, it’s to make AI usable across the business.
Governance loop
Governance determines who approves investment, how risk is assessed and how quickly resources are deployed. But the buying cycles and lock-in period for key digital systems can often stretch for multiple years – during which time AI will go through multiple capability cycles.
The risk for enterprises is that compliance and procurement processes intended to manage risk become a major drag on execution. Yet governance remains essential in highly regulated industries.
“The underlying foundation of data, access, security, compliance and risk is table stakes, but it’s a requirement before you deploy AI systems at scale,” continues Goodman.
A more scalable governance approach starts with clearer decision rights and defined AI risk frameworks that let the business move in alignment at speed.
ROI loop
While nine in ten enterprises we surveyed report seeing some value from modernization initiatives, more than six in ten still say they have not yet reached optimal outcomes. In short, progress is currently real but uneven. And when value remains isolated, AI expansion proceeds cautiously. IT leaders should encourage ROI to be evaluated enterprise-wide, rather than inside individual projects.
Path to progress
Successful AI programs are not driven by models alone. They’re driven by reinforcing systems beneath them – starting with the network foundations. Enterprises with advanced foundations are nearly twice as likely to report high business value from AI.
From there, IT leaders can build on the other loops that turn AI ambition into execution. “Good infrastructure is invisible,” says Rajat Gopal, Vice President, Cloud Networking and Security Solutions at Tata Communications. “The network should function like a utility grid. Developers simply declare their intent and the network fabric will automatically provision the path, secure it and optimize cost without human intervention.”
When these forces move together, AI becomes embedded in how the enterprise runs, and gains are enduring rather than episodic.
AI-powered enterprises are being built today. It’s time to get real about your AI readiness. Discover how to evolve your network for the next era in Tata Communications latest whitepaper.
