AI in the flow of work: Why real opportunity lies in the execution layer

AI transformation: Why enterprise leaders must focus on the execution layer

As organisations race to integrate AI into their everyday operations, many leaders are still grappling with a fundamental question: where does AI create the greatest value?

That question sat at the heart of a recent HotTopics C-Suite Exchange, AI in the Flow of Work: The Rise of the Execution Layer, held in partnership with Zoom. Bringing together senior technology and business leaders, the discussion explored how AI is moving beyond just information retrieval and task automation to become more of an active participant in business execution.

The conversation highlighted somewhat of a growing divide between organisations chasing short-term productivity gains and those redesigning processes to unlock long-term business outcomes. It also raised some important questions around trust, governance, leadership, employee wellbeing and the skills required to successfully navigate the next phase of enterprise AI adoption.

The effectiveness advantage

For many organisations, AI initiatives begin with a familiar objective: efficiency.

Reducing costs, accelerating workflows and automating repetitive tasks are often the first use cases presented to boards and executive teams. Yet according to NED and Executive Advisor, Benedetto Conversano, organisations focused exclusively on efficiency may be missing the larger opportunity.

“The trade-off between efficiency and effectiveness in the implementation of AI” is becoming increasingly apparent, he argued, noting that many organisations are struggling to generate meaningful returns from initiatives focused solely on doing existing tasks faster.

“What I am observing is that these companies are the ones who are having the highest level of challenges in getting return on the investment.”

The problem, he suggested, is that many organisations simply accelerate existing processes without questioning whether those processes remain fit for purpose.

“The companies that are doing this in the best way are the ones that re-engineer the process.”

The distinction between efficiency and effectiveness emerged as one of the strongest themes of the session. Rather than using AI to shave minutes off from existing workflows, leading organisations are redesigning workflows altogether, using process intelligence and process mining capabilities to identify friction, improve decision-making and achieve outcomes that were previously difficult or impossible.

Conversano’s examples ranged from healthcare applications designed to improve patient outcomes to consumer goods organisations using AI to transform portfolio investment decisions. In each case, success was measured not by speed alone, but by improved business outcomes.

“My advice is to go for effectiveness. It pays much better.”

Why business transformation matters more than technology

A recurring theme throughout the discussion was the danger of treating AI as a technology project rather than a business transformation initiative.

Executives continue to face pressure from boards, investors and industry commentators to demonstrate AI progress. On the other hand, Conversano warned that this external pressure often distracts organisations from addressing the underlying business challenges they are trying to solve.

“The most powerful industry in the world, the technology industry, has created another product that is going to make a lot of money. What do they want to do? Sell it as much as possible.”

He also highlighted the role of consultants, media narratives, and board-level pressure in creating what many leaders recognise as AI-driven fear of missing out. His antidote is remarkably straightforward:

“What's your business strategy? Can you make your business strategy stronger in a new context where there are new opportunities for technology?”

Only after answering that question should organisations focus on specific AI implementations. “And if the answer is yes, then the second question is, which problem are you going to solve you could not solve before?”

This approach reframes AI overall from a technology-first conversation into a business value conversation, helping organisations prioritise investments that support strategic outcomes rather than chasing technology trends.

Trust: the hidden accelerator of AI adoption

While much of the public discussion around AI centres on technology capabilities, the exchange identified another factor that may ultimately determine success or failure: trust.

“The second point I want to share with you is about what is enabling and what is creating the biggest barrier in the adoption of AI.”

The answer, Conversano argued, is often leadership credibility.

“There is one common denominator that is counterintuitive, or at least people don't talk about that much, and is a level of trust in leadership.”

According to his observations, organisations with higher levels of trust tend to experience smoother AI adoption journeys. Employees are more likely to embrace change when they believe that leaders are acting in the interests of the organisation and its people, rather than pursuing cost reduction - at any cost.

“With higher trust, I believe that if leadership comes to me and is encouraging user AI, it's not because they have one agenda which is to remove my job.”

The discussion reinforced a broader truth about digital transformation, technology adoption is rarely a technology problem. More often, it is a leadership challenge. “At the end of the day, it boils down to leadership and purpose.”

AI literacy is becoming a leadership requirement

The pace of AI innovation continues to create anxiety among organisations trying to keep up with rapidly evolving technologies.

Yet Conversano argued that executives cannot afford to remain mere passive observers: “AI is not a technology that you can learn reading a magazine on a gazette or a 15 minutes block.”

Drawing on his own experience, he described undertaking executive education focused specifically on AI and encouraging senior members of his team to do the same. “When they came back, everybody said, now we got you.”

The lesson is clear: AI literacy is becoming a core leadership competency. More importantly, he cautioned against confusing familiarity with understanding.

“There was a time when I was only reading. And I understood that, you know, it was just smoke in the eyes. You make an illusion of yourself.”

Instead, leaders need a structured education and practical experimentation to build a more meaningful understanding. Conversano noted: “I have put in my agenda half a day a week that I devote to keep myself relevant and updated.”

Governance, guardrails, and the human impact of AI

As AI becomes increasingly embedded in day-to-day operations, organisations face a dual challenge: maintaining effective governance while ensuring technology enhances rather than diminishes the employee experience.

One of the defining characteristics of generative AI is its accessibility. Unlike previous waves of enterprise technology, AI tools are readily available to almost anyone in an organisation, creating unprecedented opportunities for innovation and experimentation.

“AI is one of the most democratic technology ever available,” Conversano observed. “Everybody can use it.”

That democratisation is both a strength and a risk. While it lowers barriers to entry and accelerates adoption, it also increases the likelihood of uncontrolled proliferation, shadow IT, inconsistent practices and poorly governed use cases. Several leaders highlighted concerns around AI governance, compliance and oversight as employees increasingly experiment with new tools across the organisation.

In Conversano’s view, governance must be established from the outset.

“The power of AI shows that if you now don't have governance, you are going to make a really big mess.”

However, governance extends beyond regulatory compliance, data management and risk controls. It also requires organisations to consider how AI affects people using it. One of the more surprising insights from the discussion centred on employee wellbeing, particularly as AI becomes more deeply integrated into individual workflows.

“When organisations are pushing the use of AI at individual level, that is a correlation with higher burnout,” Conversano warned.

While AI is often positioned as a productivity enhancer, he argued that excessive individual use can create unintended pressures, encouraging employees to work faster, longer and more intensely. By contrast, organisations that deploy AI at a collective level to remove friction from business processes are more likely to realise both performance and wellbeing benefits.

The rise of the execution layer

The discussion ultimately returned to the event’s central theme: the rise of the execution layer.

The next phase of enterprise AI is not simply about generating content, summarising meetings or asking questions. It is about embedding intelligence directly into business processes, decision-making frameworks and operational workflows.

Success will not be determined by how many AI pilots an organisation launches, nor how many tools it deploys. It will depend on whether leaders can connect AI investments to strategic business outcomes, build trust, strengthen governance and redesign processes around effectiveness rather than efficiency alone.

As Conversano concluded, organisations should avoid starting with technology and instead focus on the problems they are trying to solve.