EY: Digital Transformation Drives Shift to AI-Led Operations

The integration of AI into enterprise operations is set to reach an inflection point in 2025, as organisations transition from experimental AI deployments to comprehensive digital transformation strategies. This shift marks a departure from traditional automation, moving towards systems where AI agents work autonomously alongside human employees.

However, these changes can bring structural challenges for global enterprises. Existing hierarchical management models face pressure to adapt as AI systems take on decision-making roles previously reserved for middle management. Meanwhile, the emergence of quantum computing capabilities introduces new cybersecurity vulnerabilities, forcing organisations to reconsider their approach to data protection.

Regulatory frameworks, meanwhile, are struggling to keep pace with technological change. The EU AI Act establishes initial governance structures for AI deployment in Europe, yet questions remain about implementation across global supply chains and data networks. This regulatory uncertainty coincides with growing pressure on enterprises to demonstrate responsible AI deployment while maintaining competitive advantages.

EY platform demonstrates future of enterprise operations

The shift towards AI-integrated operations requires new organisational structures, moving away from traditional hierarchical workflows towards objective-focused strategies. “The vision we are moving towards is one where humans interact with AI agents, while the C-suite oversees decision-making and becomes the control tower of this new hybrid operating system,” says Beatriz Sanz Saiz, EY’s Global AI Sector Lead.

EY tests this transformation through its EY.ai platform, which brings together human capabilities and AI to help clients transform their businesses through confident and responsible adoption of AI. “At EY, we are already navigating this transformation through our EY.ai platform. This initiative showcases how a flexible, adaptive approach to AI integration can revolutionise work structures and outcomes,” says Beatriz.

The platform operates under a “Client Zero” model, where the firm implements technologies internally before client deployment. This approach enables EY to demonstrate how AI systems integrate with existing business processes and establish foundations for wider enterprise adoption.

Industry transformation patterns emerge

Healthcare and financial services are set to lead AI implementation due to established regulatory frameworks and strategic importance. “In health, AI will enhance diagnostics, personalise treatment plans and streamline administrative processes,” says Beatriz.

The education sector faces implementation barriers despite technological readiness. “Although this is one of the sectors with the highest potential for being transformed by AI, it’s likely that an initial lack of commitment and investment will make this transformation take longer than anticipated,” says Beatriz.

“The enterprise sector will see significant changes, with AI automating routine tasks and boosting productivity,” she adds. “Sectors such as software engineering will benefit from AI reducing barriers to entry, allowing more individuals to enter these fields. AI is also expected to promote social equity by closing the skills gap between high-performing and low-performing workers, driving inclusive growth across various industries.”

Organisations balance innovation with responsibility

The implementation of AI systems requires careful consideration of ethical implications. “Organisations are placing greater emphasis on ethical AI practices by embedding transparency, accountability and governance into their strategies,” says Beatriz. Regulatory frameworks, such as the EU AI Act, reflect increasing focus on aligning AI development with societal values.

This balance becomes critical as technology capabilities expand. “As we move toward AGI, maintaining robust oversight and governance becomes even more vital. Collaborative discussions among companies, policymakers and developers are crucial to achieving the right balance between advancing AI innovation and safeguarding public trust,” says Beatriz.

Enterprise transformation drives workforce evolution

The integration of AI systems also creates demand for new specialist roles. “The demand for expertise is shifting toward emerging roles such as knowledge engineers, AI ethics specialists and governance professionals,” says Beatriz. “Eventually, the new workplace – almost regardless of industry – will be one in which humans will interact with AI agents in some way. AI Agents will be pervasive across the enterprise.”

Generative AI (Gen AI), which creates content from text prompts, reduces barriers to technical roles. “In the enterprise space, Gen AI is expected to revolutionise work by shifting from workflow-driven technology to objective-driven technology. Low-skilled workers will benefit significantly, as AI tools will enhance their productivity, enabling them to operate alongside more skilled workers.”

The transition requires significant investment in workforce development. “To bridge this skills gap, upskilling and reskilling programmes are becoming indispensable. Companies need to equip their workforce with the tools and confidence to succeed in AI-integrated environments,” says Beatriz.

AI systems democratise access to technical capabilities. “AI is simplifying complex tasks, such as coding, enabling mid-skilled workers to engage in higher-level contributions. This democratisation of skills is driving inclusivity and helping to address productivity disparities across industries.”

Security challenges reshape corporate defence strategies

Digital transformation creates new vulnerabilities in corporate networks. “The rise in the general availability of AI creates a new threat – from the ability of attackers to use Gen AI to create deepfakes and more advanced malware, to a new data risk from the creation of large and small language models, which contain sensitive data and could be subject to data poisoning,” comments Richard Watson, EY Global and Asia-Pacific Cybersecurity Consulting Leader.

The emergence of quantum computing presents additional security concerns. “2025 will be the year in which the cyber risk posed by quantum computing will hit the corporate radar, and the beginning of the development of strategies to mitigate a quantum risk that will inevitably grow over time,” he says.

IoT devices present growing security risks. “As we become ever more dependent on IoT devices, their resilience to cyber attack will be a growing concern,” says Richard.

Despite creating new vulnerabilities, AI strengthens security capabilities. “AI creates opportunities for cyber defense. These include the automation of tasks that are highly data intensive, such as threat detection and response, vulnerability management and identity and access management. This allows organisations to get more coverage and free up cyber professionals from laborious data analysis, so they can focus on higher-order risks and insights,” says Richard.

Organisations develop comprehensive security frameworks

Companies require specific strategies for AI security. “Organisations need to do five things: First, develop a specific cyber strategy for AI – one that prioritises mean time to detect and respond. Second, consider all your vulnerabilities, on premise, in the cloud and in the supply chain. Third, make sure you invest more and more in the culture of cyber – helping those across the business who are using systems and data to be aware of the data they are using and what the risks might be to it, using it in AI scenarios,” says Richard.

He continues: “Fourth, organisations need to simplify both their cyber technology and broader technology landscapes, to make the task of defending systems, data and networks easier. And fifth – last but by no means least – organisations need to think about their investment in cyber as a value creation activity, not just as a cost of doing business.”

Corporate training evolves with technology

Traditional security training approaches become obsolete. “A risk based, data-driven approach to cybersecurity training and awareness is now possible – with tailored, experiential training, customised to the individual’s role, the data and systems they have access to, an approach which allows the training to be defined by the specific ways the person being trained actually uses the systems concerned,” says Richard.

The integration of AI into security products remains uncertain. “One thing of particular interest i will be watching how Gen AI is integrated into more and more products,” Richard explains. “In terms of both use cases and licensing model, it largely remains to be seen exactly how this will play out. The lack of clear direction to date has caused an impediment to the take-up of Gen AI in cyber defence, so I hope and expect to see real progress in this area during 2025.”

To read the full story in the magazine click HERE

Explore the latest edition of Data Centre Magazine  and be part of the conversation at our global conference series, Tech & AI LIVE and Data Centre LIVE. 

Discover all our upcoming events and secure your tickets today. 

Data Centre Magazine is a BizClik brand