How AI and Regulatory Compliance Shape Digital Infrastructures

The demands placed on digital infrastructure have never been greater. Omdia forecasts that, by 2030, a staggering three-quarters of network traffic will be infused with AI, fuelled by continual growth in video and image content. This new breed of content is fundamentally different, requiring real-time responsiveness, lossless delivery, and heightened sensitivity to privacy. The infrastructure supporting these workloads must not only be robust and scalable, but also agile enough to handle the unique challenges posed by AI workloads. 

Modern infrastructure is now an essential bedrock for AI solutions, spanning cloud services, connectivity, and cybersecurity. It forms the first link in a complex AI value chain supporting everything from third-party AI platforms to the emergence of agentic AI, where autonomous agents perform tasks in support of humans. For instance, Gartner predicts that, within the next few years, AI agents will resolve the majority of routine customer queries, underscoring the magnitude of this shift. 

However, as organisations embrace AI, the stakes for data privacy and sovereignty have risen sharply. Sensitive information now traverses global networks with increasing frequency, forcing businesses to grapple with heightened risks and regulatory scrutiny. Compliance pressures are compounded by landmark legislation such as the U.S. CLOUD Act and the EU’s Data Act, which have made it imperative for businesses to know exactly where their data resides and how it is processed. 

This new landscape is prompting widespread strategic re-evaluation. A recent survey by Global Data, commissioned by Orange Business, found that 70% of enterprises agreed that they needed to re-evaluate their cloud strategy due to GenAI, 43% said the same about the need to modernize their network infrastructure, and 63% said that GenAI required a re-evaluation of their cybersecurity strategy. Yet, despite these efforts, Gartner estimates that at least half of GenAI projects will be abandoned after proof of concept, citing poor data quality, inadequate risk controls, and unclear business value as primary culprits. The challenges are only set to intensify, with more than 40% of agentic AI projects predicted to be cancelled by 2027 for similar reasons. 

To thrive in this demanding environment, digital infrastructures must be flexible, adaptable, and highly transparent. The ability to dynamically configure network services for diverse use cases, while embedding governance policies that address data sovereignty and jurisdictional compliance, is now non-negotiable. Organisations that achieve this not only gain technical agility, but also the assurance that their most sensitive data remains shielded from extraterritorial legal claims. This, in turn, strengthens digital trust among both customers and stakeholders. 

Many businesses are increasingly turning to sovereign data centres within their own legal jurisdictions to mitigate compliance risks and protect against unwanted external access. Trust is built on a combination of technical resilience, ethical stewardship, and transparent governance. Persistent challenges such as cyber threats, data breaches, and the potentially opaque decision-making of AI systems (“black box” AI) have made it clear that transparency and accountability are vital. Where responsibility is diffused or governance mechanisms are unclear, trust quickly erodes. As AI-driven proof-of-concepts move into full operational deployment, the need for clear lines of accountability and robust governance becomes ever more critical. 

As a recognised leader in connectivity and cloud services, Orange Business brings decades of infrastructure expertise to the table. The company has a proven track record of integrating AI into network operations and cybersecurity, including the use of AI for root cause analysis and agentic incident triage, as well as the development of Live Intelligence, its trusted AI platform, built in partnership with LangChain and integrating models from giants such as Google, OpenAI, Anthropic and Mistral. Today, this platform is used by 70,000 employees every month, as well over 150 customer companies. The platform natively supports trust needs: it uses our private cloud infrastructure in France, proposes on-prem, open-weight models as well as leading frontier models, and ensuring that clients keep full control over their choice of cloud and LMM partners and avoiding lock-in concerns. 

Orange Business also enables organisations to deploy agentic AI solutions across a spectrum of business functions, from Customer Experience (resolving queries) to operations (predictive maintenance and inventory management), and even employee experience (content creation and pricing optimisation). Its Evolution Platform empowers customers to adapt to changing AI-driven workloads, while its focus on security helps defend against the growing sophistication of cyber threats. 

In summary, the growing demands of AI and the shifting landscape of digital trust are reshaping the very foundation of digital infrastructure. The move towards sovereign data centres and jurisdiction-aware architectures will define the next era of digital transformation. For organisations seeking to harness the true potential of AI without compromising on trust, security, or compliance, the choices they make about their digital infrastructure today will have a profound influence on their success in the years to come.