As AI adoption becomes widespread, competitive advantage is shifting from access to how effectively organisations turn AI into lasting capability. Fahed Bizzari argues that the real differentiator lies in capturing employee expertise, redesigning work and building systems that allow successful AI use to spread across the organisation.

[For more news, click here]

Simply adopting AI is not enough to distinguish one organization from another, it is what they do with those capabilities that matter. In discussion with Tech Revolt, Fahed Bizzari, Founder of Bellamy Alden highlighted that AI adoption itself is a weak measure of competitive advantage and further spoke on how organizations can turn individual AI expertise into shared systems and create real value over time.

Q1. The UAE has pushed AI adoption faster than almost anywhere else. Once access and experimentation become widespread, does “AI adoption” stop being a meaningful competitive advantage? What begins to separate leading organisations at that point?

To be frank, AI adoption has never been a meaningful measure of competitive advantage. Adopting AI simply means people use it. It tells us nothing about how safe, effective or judicious that use is or the type of work it's applied to. Being early adopters can give you a head start, but you still have to do something worthwhile with it.
The real advantage comes when teams harness AI's power to expand what they can do, which is what I call “AI empowerment”. They learn to use it safely, effectively and judiciously in work that matters so that they can then do existing work better and faster or take on work that was previously beyond them.

For an organisation, the challenge is to establish that individual empowerment as a lasting organisational strength. For example, whilst delivering a workshop to 30 managers, I met a participant who had built an impressive AI assistant that made her work better and easier. But when I asked who else was using it, the room fell silent. What separates leading organisations is that they recognise discoveries like hers, assess where else they could help, and support other people in using them.

Q2. You have argued that many organisations confuse using AI with actually harnessing its power. What does an organisation that has genuinely restructured around AI look like compared with one where employees are simply using Copilot, ChatGPT or other tools more frequently?

First, I think those two things can overlap because, used well, tools like Copilot and ChatGPT can drive the kind of restructuring you're asking about. For example, the marketing team of one company we worked with built an AI assistant using technical documentation that now saves them from having to get information from engineers. Their engineers still review the marketing material they produce, but with far fewer interruptions and less waiting time.

As for how organisations look when they restructure, it really depends on whether the function or department in question is AI-first or human-first. To explain, human-first is where the human does the operational work and the AI assists them. Sales, for example, is a very human-first function because someone still needs to hold the client relationship and lead the conversations. When they then use AI for research, meeting preparation, proposal writing and other sales-related activities, they find themselves needing less support from others. The work of those others then needs to be redirected elsewhere.

As for AI-first, that's when AI takes on the core operational work. For example, monitoring bid portals and evaluating how well-matched every opportunity is can now be done almost entirely by AI. But to do so, the AI needs people to build it, monitor it, improve it and handle scenarios where it falls short. That, of course, changes the roles and skills you need to run that operation. A great place to see this change in people's roles is at the Abu Dhabi Autonomous Racing League (A2RL), which I went to watch in 2025. As an F1 fan, I found it wild to see AI race the cars and engineers stand on the podiums.

Whether it's AI-first or human-first, leaders need to recognise when harnessing AI's power changes the actual work and then update job descriptions, staffing and team structures to reflect how the work now happens.

Q3. You've described how AI can change people's roles and the way teams work. What capabilities do organisations need to develop in their workforce to make that possible?

To begin with, everyone needs to learn the two fundamentals of harnessing AI. The first is knowing how to stay safe whilst using it so that the pursuit of its benefits does not inadvertently lead to harming themselves and their companies. The second is learning how to align their AI tools with their work realities so that AI can become more valuable to them. With those foundations in place, teams can begin to connect their AI to relevant information and tools.

You also need the company's experts to go further and begin building agent skills. This is when you write out how to complete a given task so that an AI can either assist you in completing it or complete it outright. An HR professional might build a skill to prepare induction plans for different roles, while a marketing professional might build one to convert product information into a brochure.

A company that gets that far is in an incredible place and will soon feel the need to restructure based on the human-first approach I mentioned. When it eventually concludes that certain functions can become AI-first, it needs to think about engineering. That usually means involving engineers, although no-code and low-code solutions are increasingly putting engineering abilities in the hands of non-engineers.

Q4. How should organisations identify where AI could make a meaningful difference and decide which opportunities to pursue first?

The easiest and most effective way to begin is by helping people make an inventory of the documents they regularly produce and process. Every document represents work that has either gone into it or gets triggered by it, and all of that signals places where AI could make a meaningful difference.
From there, they can identify the ones that come up frequently, involve substantial work or have an important bearing on the business and prioritise according to anticipated impact.

For example, contracts came up frequently in the commercial team at one company we worked with. This led them to recognise how much work went into processing those contracts because, for deals worth less than US$2 million, team members had to review them themselves according to guidelines provided by their legal team.

That revealed an opportunity for AI to systematically check each contract against those guidelines, propose revised clauses in the contract's own writing style and draft emails explaining the proposed changes to the other party. Reviews that had taken the better part of a day came down to about half an hour, eventually saving thousands of hours a year across the regional team. When the company extends that approach to its teams elsewhere in the world, the savings will multiply further.

Q5. If two GCC companies have access to essentially the same AI models and technology, where does a lasting competitive advantage come from?

It all comes down to how they harness the technology, both at the individual and organisational levels. At the individual level, as I mentioned earlier, people must learn how to infuse their expertise into the AI by aligning it with their work realities and teaching it how to perform given tasks. For example, one of our clients has a senior employee responsible for assessing incoming claims. Over the years, he developed a detailed approach to interrogating every claim before deciding how to respond. He used examples of past claims, decisions and outcomes to help AI draw out the questions he would normally ask himself. He then used that documented method to build an assistant that could guide the assessment and flag concerns, while he still made the final decision. Although other staff in other companies have access to the same AI that he does, he's been able to harness it better.

Then, at the organisational level, the company needs the infrastructure that allows people to capture, share and maintain what they've learned or built. For example, a shared place to store and find approved agent skills, assistants and reference material. They also need to give their people a way to route feedback into those resources so they keep improving. The company also needs to provide continuous training and support that anyone can tap into, especially new joiners.

The productivity gains are valuable in their own right. But as people learn what AI makes possible, they can also begin to imagine products, services and ways of working they hadn't considered before because they seemed impossible or impractical. That opens up possibilities for the business that competitors still haven't even imagined.

Q6. How should boards assess the return on their AI investment? Are hours saved and headcount efficiency enough to capture its value?

In my early days, I found board pressure to demonstrate ROI very annoying because the value of AI seemed obvious to me. However, I've since changed my opinion and am very supportive of it because it forces teams to focus on opportunities that provide the highest ROI first. That, in turn, leads to more support from the board, especially on AI opportunities which are obviously important but where the ROI is not easy to measure. Boards should compare the benefits actually achieved with the cost of building, running and supporting the system.

As for the question of hours saved or headcount efficiency, these are definitely not the be all and end all of the value that AI can deliver. For example, AI can help people produce better work, catch mistakes and make more informed decisions. It can also help leaders achieve something they've wanted for years: systematising how work gets done so the company is less dependent on particular individuals. The senior employee I mentioned who assesses incoming claims is already semi-retired and works part-time. His departure would leave a serious gap. By capturing his approach in an AI system that others can use, the company is retaining expertise it would otherwise lose if he retired unexpectedly.

There's also the innovation side. Harnessing AI can lead a company to develop a service it couldn't previously offer, or make something affordable that used to be more expensive. The value there comes from creating new opportunities which only arise from experimentation without ever thinking about how much they are worth. That’s why I’d start with opportunities where the returns are easier to demonstrate as they give boards the confidence to invest in opportunities whose value takes longer to establish.

Q7. Vision 2030 has accelerated digital transformation across Saudi Arabia, but AI raises a deeper question: how can organisations built around an existing way of working adapt as that work changes? What must Saudi business leaders do?

The first step is for Saudi leaders to take full responsibility for AI empowerment within their organisations. They need to establish what they want the organisation to become capable of and commit the resources to help people get there. Yes, they can appoint a senior sponsor to lead that work, but they must remain accountable for its direction and progress. That's the top-down part of what I call a Top-Down Inside-Out approach.

The inside-out part comes from involving employees in the transformation and then absorbing the intelligence that teams generate into how the company works. Involving them means giving them time to learn and apply what they're learning, access to suitable tools and someone to turn to when they get stuck. Their managers also need to make room for that work in their schedules, so participation is something the company supports in practice.

As for absorbing the intelligence, that means several things. For example, when things go well, they must help that success spread to other parts of the company where it could be useful. That could mean making an agent skill available to colleagues, helping another team adopt a better way of working or incorporating what has been learned into the company's standard practices.

When they face challenges, for example with people feeling worried that sharing their expertise with AI could put their jobs at risk, they must understand what's behind those concerns and address them. What they learn should then shape how they introduce AI elsewhere, including how they explain changes to people's roles and support their development. And when the same questions keep coming up, spoken or unspoken, they must ensure the answers are easy to find.

Saudi leaders must empower their people with AI so their people can empower the organisation. By doing so, they can turn scattered AI use into a lasting organisational strength.

Where Competitive Advantage Moves Next.

Fahed Bizzari’s arguments showcase a shift in the AI conversation. The organizations that succeed will not necessarily be the ones with access to most advanced models but those that become better at absorbing what their people learn and turning that knowledge into repeatable ways of working. Leadership, skills, incentives and the ability to capture institutional knowledge could ultimately determine where AI remains a collection of useful tools or becomes a durable competitive advantage.

Related Articles

Everpure Targets the Data Bottleneck Holding Back Enterprise AI

VAST Data and NVIDIA Target the Infrastructure Challenges of Autonomous AI Agents

SUSE and NVIDIA Target the Governance Gap in Enterprise Agentic AI