A strange thing is happening inside companies.

The CEO says:

“We need an AI strategy.”

A committee is formed.

Someone creates a presentation.

Another team starts testing ChatGPT.

IT evaluates Microsoft Copilot.

Marketing buys three new AI tools.

Customer service launches a chatbot.

Someone proposes an AI policy.

Six months later, the company can proudly say:

“We are using AI.”

There is only one problem.

The company itself has barely changed.

Same processes.

Same departments.

Same approvals.

Same customer experience.

Same management structure.

Same bottlenecks.

Only now there is AI sitting on top of all of them.

That is not transformation.

That is decoration.

And it is why I think many companies are starting with the wrong question.

The question is not:

What should our AI strategy be?

The better question is:

What should our business strategy become now that AI exists?

There is a big difference between the two.

And that difference may determine which companies actually capture value from artificial intelligence—and which simply accumulate more software.

We Made the Same Mistake With Digital Transformation

This pattern isn’t new.

Twenty years ago, companies talked about their “internet strategy.”

Then their “mobile strategy.”

Then their “social media strategy.”

Then their “digital strategy.”

Eventually, the distinction became absurd.

Imagine a retailer today saying:

“We have our business strategy over here…and our internet strategy over there.”

The internet is no longer a separate strategy.

It is part of how the company sells.

How it communicates.

How it operates.

How customers discover it.

How employees work.

How suppliers connect.

How information moves.

Digital stopped being a department and became part of the architecture of business.

AI is moving in exactly the same direction—only faster.

Udjat has made a similar argument around digital transformation in Dubai: technology does not create transformation by itself. Strategy, leadership, operations and execution have to move with it.

AI makes that lesson even more important.

Because AI isn’t simply another channel.

It can affect almost every function inside the company.

Sales.

Marketing.

Finance.

Operations.

Customer service.

Product development.

Technology.

Human resources.

Knowledge management.

Management itself.

So why would something capable of changing the entire organization live inside a separate “AI strategy”?

The Problem With an AI Strategy

There is nothing inherently wrong with creating an AI roadmap.

Companies need priorities.

They need governance.

They need investment plans.

They need policies.

They need technical architecture.

They need owners.

They need a clear understanding of where AI should—and should not—be used.

The problem begins when the AI strategy becomes separate from the business strategy.

Then companies start asking questions like:

Where can we add AI?

Which AI tools should we buy?

Which department should pilot AI first?

Should we build a chatbot?

Should everyone get Copilot?

These sound like strategic questions.

Most aren’t.

They’re technology decisions.

A business strategy asks something different:

Where are we trying to win?

What value do customers choose us for?

What stops us from growing?

Where is our cost structure weak?

What takes too long?

What knowledge do we have that competitors don’t?

Which decisions determine our performance?

What could we do now that wasn’t economically possible before?

Then—and only then—you ask:

How does AI change the answer?

That’s a much more dangerous question.

Because sometimes the answer isn’t “add an AI tool.”

Sometimes the answer is:

Remove the process.

Sometimes:

Change the role.

Sometimes:

Build a completely different customer experience.

Sometimes:

Automate 80% of the workflow.

Sometimes:

Keep AI away from it entirely.

And sometimes:

Change the business model.

That is why an AI strategy cannot live comfortably beside the business.

Eventually, it has to enter it.

AI Is Not a Project

This is probably the first mindset companies need to abandon.

AI is frequently managed as a project.

Project starts.

Project has a sponsor.

Project has a budget.

Project deploys a solution.

Project ends.

But what happens when the model improves three months later?

What happens when agents gain new capabilities?

What happens when an activity that required five employees last year can suddenly be performed by one employee supervising several agents?

What happens when competitors redesign a customer journey that you spent ten years optimizing?

What happens when a process that previously cost AED 100 per transaction can be performed for AED 10?

You don’t have a project anymore.

You have a change in the economics of the business.

And business strategy exists precisely because economics change.

Brightery’s work around AI transformation makes an important distinction here: deploying individual tools is not the same thing as transforming how a business operates.

That’s the distinction boards and CEOs increasingly need to understand.

There Are Three Levels of AI Adoption

One reason companies struggle with AI strategy is that very different activities are being grouped under one phrase:

“We’re adopting AI.”

That could mean almost anything.

A useful way to think about it is through three levels.

Level 1: AI-Assisted Work

An employee does the job.

AI helps.

A marketer drafts content faster.

A developer gets coding assistance.

An analyst summarizes a report.

A salesperson researches a prospect.

Useful?

Absolutely.

Transformation?

Not necessarily.

The company still operates roughly the same way.

Level 2: AI-Enabled Processes

Now AI enters the workflow.

An inquiry arrives.

AI classifies it.

Data is retrieved.

A response is generated.

The CRM is updated.

A salesperson is alerted if the opportunity passes a certain threshold.

The employee is no longer merely “using AI.”

The process has changed.

This is where areas such as AI marketing automation in the UAE become interesting. The value isn’t simply that AI can write an email. The value comes when customer data, decisions, automation, follow-up and human intervention become part of one connected system.

Brightery describes the same operational shift in its work on how businesses can automate company operations.

Now we are getting closer to transformation.

But there is another level.

Level 3: AI-Native Business Design

This is where the uncomfortable questions begin.

Instead of asking:

“How can AI improve this process?”

you ask:

“If we were designing this company today, knowing what AI can do, would this process exist at all?”

Would we have the same departments?

Would information travel through the same people?

Would seven approvals still be necessary?

Would every customer receive the same service?

Would managers spend hours preparing reports?

Would analysts manually assemble dashboards?

Would salespeople search through CRM records?

Would every new employee require the same administrative support?

Would customers have to wait until 9 a.m. for someone to respond?

Would this position exist?

Would this product exist?

Would this pricing model exist?

Now AI is no longer being applied to the company.

The company is being reconsidered because of AI.

That is the territory we will explore throughout The Company After AI.

Start With the Bottleneck, Not the Model

There is a useful discipline I think every leadership team should adopt.

Stop beginning AI conversations with the technology.

Don’t begin with:

“Can GPT do this?”

“Can we build an agent?”

“Should we use Gemini?”

“Should we buy Microsoft Copilot?”

Begin with:

What is preventing this company from performing better?

Maybe customers wait too long.

Maybe salespeople spend 40% of their time performing administrative work.

Maybe your company possesses years of valuable knowledge but employees cannot find it.

Maybe management decisions require five reports and three meetings.

Maybe marketing produces hundreds of leads but sales cannot prioritize them.

Maybe every employee performs the same repetitive data-entry work.

Maybe your CRM, ERP, website, finance platform and communication systems barely talk to each other.

Maybe the problem isn’t AI at all.

That’s important too.

Because a mature AI strategy sometimes reaches this conclusion:

“We shouldn’t use AI here.”

AI isn’t the objective.

Business performance is.

Udjat’s existing work on AI for business in Dubai already points toward this distinction in marketing: AI is most useful when connected to skilled people, quality data and broader business objectives rather than deployed in isolation.

The same principle applies to the rest of the organization.

The Real AI Strategy Has Five Questions

If I were sitting with a leadership team today, I would not begin by asking them to list AI use cases.

I would put five questions on the screen.

1. Where Do We Create Value?

Why does the customer pay us?

What do we do better than alternatives?

Where does our knowledge matter?

Where does speed matter?

Where does trust matter?

Where does personalization matter?

Where does scale matter?

AI should strengthen—or challenge—those answers.

2. Where Does Work Get Stuck?

Every company has friction.

Approvals.

Handoffs.

Search.

Reporting.

Repeated communication.

Data entry.

Duplicate work.

Waiting.

AI has enormous potential here.

But sometimes automation merely accelerates a badly designed process.

A six-step process does not automatically become intelligent because step four contains an LLM.

Sometimes the most valuable thing you can do is delete three steps before automating the remaining three.

We will return to this in another article in this series:

Don’t Automate a Bad Process.

3. What Should Humans Do?

This may become one of the defining management questions of the next decade.

AI changes the answer to:

Who should do the work?

Today, most companies divide work between employees, contractors and software.

Tomorrow, there may be another category:

agents.

Then each workflow has to answer:

Should a human perform this?

Should AI assist the human?

Should AI perform it and ask for approval?

Should an agent execute it autonomously within limits?

Should multiple agents coordinate the process?

And where must human accountability remain?

That’s organizational design.

Not merely AI adoption.

4. What Does AI Need to Know?

An AI model with no access to your company’s knowledge is impressive—but limited.

It may understand accounting.

It does not automatically understand your accounting.

It may understand customer service.

It does not automatically know:

your customers,

your policies,

your contracts,

your exceptions,

your historical decisions,

your inventory,

your CRM,

your standards,

your pricing,

or why your company does certain things differently.

This is one reason the conversation eventually moves from AI tools toward infrastructure, integration and company data.

Brightery’s work on AI-empowered software is relevant here because useful enterprise AI increasingly has to connect intelligence with the systems where business actually happens.

The question isn’t only:

How intelligent is the AI?

It’s:

What does the AI know about our business—and what is it allowed to do with that knowledge?

5. How Will We Know It Worked?

This may be the question companies avoid most.

AI produces impressive demonstrations.

Demonstrations are not ROI.

The employee says:

“It saves me two hours.”

Good.

Then ask:

What happened to the two hours?

Did revenue increase?

Did output increase?

Did headcount requirements change?

Did customer response time fall?

Did error rates improve?

Did conversion increase?

Did the employee perform more valuable work?

Did the company actually become faster?

Or did we simply create two extra hours that were absorbed by more email, more meetings and more work?

An AI strategy without economics is a technology wish list.

A business strategy must eventually answer:

Where is the value?

The UAE Has a Particular Opportunity

There is another reason this conversation matters in the UAE.

The country isn’t waiting to see whether AI becomes important.

The UAE National Strategy for Artificial Intelligence 2031 explicitly aims to establish the country as a global leader in AI and identifies sectors including resources and energy, logistics and transport, tourism and hospitality, healthcare and cybersecurity as priority areas for transformation. [1]

That creates an unusually ambitious environment.

And Middle East CEOs appear to share that ambition.

PwC’s 2026 Middle East CEO research found that 70% of regional CEOs say their organizations already have a clearly defined roadmap for AI initiatives.

But another number is more interesting.

Only 22% said their most-used AI tool has access to all relevant company documents and data. In the GCC, the figure was only 16%. [2]

Think about that contradiction.

The roadmap exists.

The ambition exists.

The investment exists.

But the AI may still be standing outside the company’s actual knowledge.

This is exactly why strategy cannot stop at:

“Adopt AI.”

The difficult work begins after that decision.

The Company Is the Product That Needs Redesigning

Microsoft’s 2026 Work Trend Index contains one observation that I think executives should pay particular attention to.

Its research found that organizational factors—things such as culture, management support and talent practices—were associated with more than twice the reported AI impact of individual factors. [3]

That’s significant.

Because companies have spent enormous energy teaching employees:

How to prompt.

How to use Copilot.

How to generate content.

How to summarize documents.

How to automate small tasks.

Those skills matter.

But imagine installing a Formula 1 engine inside a car whose wheels are locked.

Employee capability can move quickly.

The organizational system around the employee can remain painfully slow.

Old approvals.

Old incentives.

Old reporting structures.

Old job descriptions.

Old software.

Old permissions.

Old KPIs.

Old management assumptions.

Then leadership wonders why the AI investment isn’t transformational.

It is because we changed the worker.

We didn’t change the work.

Microsoft calls the leadership challenge “rearchitecting work.” [3]

I think the idea goes further.

Eventually, we have to rearchitect the company.

So Do You Need an AI Strategy?

Yes.

For now.

But only in the same way companies once needed an internet strategy.

You need a focused effort because the technology is new.

You need people exploring it.

You need standards.

Governance.

Investment.

Architecture.

Experiments.

Education.

Priorities.

But the success of the AI strategy should eventually make the AI strategy disappear.

Because AI becomes embedded into:

business strategy,

operations,

product development,

workforce planning,

customer experience,

technology,

capital allocation,

risk,

and management.

Deloitte’s 2026 Global CSO research captures this shift clearly: AI can no longer be treated merely as a technology agenda adjacent to corporate strategy. Its implications increasingly belong inside planning, capital allocation, operating-model decisions and talent strategy. [4]

That’s the destination.

Not:

Business Strategy + AI Strategy

But:

Business Strategy in a World With AI

A Better Conversation for Your Next Leadership Meeting

Instead of asking your leadership team:

“What is our AI strategy?”

Try these questions.

Strategy

What competitive assumption has AI changed in our industry?

What can a competitor now do that was economically impossible three years ago?

What can we now offer customers that wasn’t previously viable?

Operations

Which workflows consume the most human time?

Which processes exist because information used to be difficult to access or process?

Which steps could disappear rather than simply become faster?

People

What work should remain human?

Which roles should be augmented?

Which tasks should be automated?

Where must human accountability remain?

Data

What does AI need to know about our company?

Can it access that information reliably?

Is the information structured?

Is it trustworthy?

Technology

Which systems must AI interact with?

What should we build?

What should we buy?

What needs integration?

Governance

What should AI be allowed to recommend?

What should it be allowed to execute?

Who is accountable when it gets something wrong?

Economics

Where exactly should AI create value?

What metric should change?

By how much?

By when?

If your leadership team cannot answer those questions yet, that’s fine.

But answering them is considerably more useful than collecting another list of AI tools.

This Is the Beginning of “The Company After AI”

This article isn’t arguing that companies should stop thinking strategically about artificial intelligence.

Quite the opposite.

It is arguing that AI has become too important to remain inside an AI strategy.

And that creates a much bigger set of questions.

What happens to the organizational chart?

What happens to managers?

What happens to entry-level jobs?

What happens when companies employ hundreds of AI agents?

What happens to software?

What happens to cost structures?

What happens to competitive advantage when every company has access to similar models?

What should remain human?

Who is responsible for an autonomous agent?

How do you measure AI ROI?

When should you build your own AI?

And perhaps the most important question:

If you were creating your company from zero today—with everything we now know AI can do—would you design the same company?

For most businesses, I suspect the answer is no.

And that’s where the real AI transformation begins.

What’s Next in The Company After AI

02 — Stop Asking “Where Can We Use AI?”

The next article explores a mistake hiding inside most AI transformation programs: companies take existing processes and look for places to insert AI.

We’ll reverse the question.

Instead of asking where AI fits into the company we already built, we’ll ask:

What would we build differently if AI had always existed?

Related Udjat Insights

Continue exploring the subject through Udjat’s existing research and services:

• Digital Transformation Agency in Dubai: Why Smart Businesses Choose Strategy Before Technology
• How Can AI Change Marketing for Business in Dubai?
• AI Marketing Automation UAE
• AI Marketing Services Dubai
• Generative Engine Optimization (GEO) at Udjat UAE

Related Brightery Insights

For businesses exploring the technology and implementation side of this transition:

• The Best AI Transformation Agency
• Automate Company Operations
• AI-Empowered Software: Redefining Smart Business Solutions

Sources

[1] UAE National Strategy for Artificial Intelligence 2031 — UAE Office for Artificial Intelligence. The strategy sets the UAE’s ambition to become one of the world’s AI leaders by 2031 and identifies priority areas for AI transformation.

[2] PwC — 29th Global CEO Survey, Middle East Findings, 2026. PwC reports that 70% of Middle East CEOs have a clearly defined roadmap for AI initiatives, while only 22% say their most-used AI tool has access to all relevant company documents and data; the GCC figure is 16%.

[3] Microsoft — 2026 Work Trend Index Annual Report. Microsoft surveyed 20,000 AI-using knowledge workers across 10 markets and analyzed Microsoft 365 signals. The report argues that leaders need to rearchitect work and reports that organizational factors such as culture, manager support and talent practices account for more than twice the reported AI impact of individual factors.

[4] Deloitte — 2026 Global Chief Strategy Officer Survey. Deloitte argues that AI should move beyond being a technology-side agenda and become integrated into corporate strategy, including planning, capital allocation, operating-model choices and talent strategy. The survey also found that only 16% of respondents reported using AI to fundamentally reimagine lines of business or create new sources of competitive advantage.