Inc. Arabia: If Women Are Using AI, They Should Be Helping Build It Too

Leena Salman

There is a lot to feel optimistic about right now. According to the 2026 Artificial Intelligence (AI) Index Report from Stanford’s Institute for Human-Centered AI, Saudi Arabia ranks highest in the world for empowering women in AI, with women making up 32.3 percent of AI inventors and authors, ahead of countries like Australia and Canada.

That is not a fluke. Women across this region have been shaping the digital economy for years. Wamda has reported that the share of women digital entrepreneurs in the Middle East reached 35 percent, compared with just 10 percent worldwide. Recent research also shows women across the Middle East are embracing AI at impressive rates. We are using it, adopting it, and folding it into how we work and run businesses.

So, here is the question I keep coming back to as an engineer: if women are increasingly adopting AI, are we also helping shape the technology behind it?

From Using AI to Building AI

Getting more people to use AI is good, but it is only one part of the equation. Using a tool and deciding how that tool gets made are two very different things. You can be fluent in a product and still have no say in what it does, who it serves, or what it quietly assumes about you.

The next phase of AI in this region should not only be measured by adoption numbers. It should be measured by how many women are designing, engineering, and influencing the systems that millions of people and businesses will eventually depend on. Adoption reflects who is in front of the screen today. Design shapes what is possible for everyone who comes after.

I say this as someone who lived that shift. I started as a frontend developer, moved into full-stack work building secure, real-time applications, and I now lead an AI platform. Somewhere along the way, I stopped simply using these tools, and I started shaping them. That move, from user to builder, is exactly what I want more women in this region to consider.

An Engineer’s View: The Technology Is Not Neutral

There is a comforting myth that AI is objective, that it simply reflects the world back to us. It does not. Every AI product is the sum of thousands of human decisions.

Someone chooses which data goes in, and which gets left out. Someone decides which problems are worth solving first, and which can wait. Someone sets the default behavior of the model, writes the prompts that shape its tone, and designs the interface that decides who feels welcome and who feels like an afterthought. None of those choices are neutral. They carry the assumptions of the people making them.

For most of history, the products people relied on were never built with everyone in mind. They were built for a set of groups, and everyone else adapted. I do not want AI to repeat that. Everything that makes me who I am is exactly what lets me keep the people who are usually left out in mind while I build. When AI overlooks those communities, I see it as a gap, a sign that the technology has not reached its full maturity yet. A tool that only works well for a default few is not finished. It is unfinished in a way we have simply learned to tolerate.

This is why diverse engineering teams produce stronger technology. It is not about fairness as a slogan. A wider range of people in the room means a wider range of blind spots get caught before they ship. The team notices the assumption that would have broken the product for a whole group of users. That is not softness. That is engineering quality.

Read More:The Gen AI Divide: Why Hype Won’t Cut It Anymore

Why This Matters Most Right Now in the GCC

This conversation feels particularly urgent in the GCC, because AI adoption is accelerating so quickly. Governments are making it central to national strategy. Businesses are integrating it into core operations. The systems we build today will become the default people live with for years.

And most AI tools still assume a Western starting point. They assume English input and English output. They assume Western names, date formats, and reading direction. They treat Arabic as something to translate into rather than something to build from.

At the Commercial Bank of Dubai, I built an Arabic content management system to meet new UAE Central Bank regulations. That experience taught me early on that localization is not a translation layer you add at the end. Arabic changes the structure, the layout direction, and the way content is authored and governed. If you treat it as an afterthought, it shows.

That lesson continued to shape how my team builds today. Our platform, Myndlab, is Arabic-first. You describe what you want in Arabic, and it builds the application in Arabic. It is not English software wearing an Arabic coat. It starts from the language and the reality of the people using it.

That is what representation looks like in practice. It is not a diversity statistic on a slide. It is a product that behaves differently, and better, because someone from the community was in the room when the foundational decisions got made. Localization, done properly, is not about language at all. It is about whether the technology actually reflects the people and the places it is meant to serve.

Diversity Is a Competitive Advantage, Not a Quota

It is easy to frame representation purely as the right thing to do. It is that, but framing it only that way undersells it. Diversity on an engineering team is a commercial advantage.

The perspectives around the table decide which problems get solved, which data gets trusted, and which users get designed for. Homogeneous teams ship blind spots. Diverse teams ship products that more people actually trust and adopt. In a region as linguistically and culturally specific as the GCC, the team that understands the local reality will build the product the market actually wants.

Organizations building AI should treat diversity as strategy. It improves innovation, because more kinds of problems get seen. It builds trust, because more kinds of people feel considered. And trust is what drives long-term adoption, which is the whole game.

So, the practical asks are simple. Hire differently. Fund localization as core work rather than an afterthought. And put women on the teams making the foundational decisions, not just reviewing them at the end.

The Future Is Decided by Who Is Building

The future of AI will not be defined by increasingly powerful models alone. It will be defined by the people building them, the perspectives they bring, and whether the technology genuinely reflects the needs of the communities it is meant to serve.

To any Arab woman reading this and wondering whether AI is for her: you are not just a user of this technology. You are exactly the person it needs to build it. The gaps you notice, the things that feel off, the assumptions that do not fit your life or your language, those are not reasons to stay out. They are the reason to step in. Do not wait until you feel completely ready. Apply, build the thing you wish existed, and say the thing the room is missing. AI will not reflect your community until people from it are the ones shaping it.

The good news is that this region has already proven women will show up. Now, the invitation is bigger. Not just to use what gets built, but to build it. If we want the next generation of AI to be both genuinely innovative and genuinely inclusive, we need more of us in the room where it is made, and we need more collaboration to get us there. If women are using AI, we should be helping build it too.

Originally published in Inc. Arabia