Why a Saudi AI founder thinks local enterprises need AI built for the Kingdom

Saudi organizations are still at the very early stages of AI, says Badr Al Malluh, and the global platforms on offer don’t meet the Kingdom’s regulatory requirements. After years in planning and supply chain at Mars, Pfizer, and Panda, Al Malluh co-founded Gaia to go after that gap and has raised a USD 1.5 mn pre-seed round to do it.

Badr Al Malluh, co-founder and CEO of Gaia: Each week, My Morning Routine looks at how a successful member of the community starts their day — and then throws in a couple of random business questions just for fun. Speaking to us this week is Badr Al Malluh (LinkedIn), co-founder and CEO of Gaia. Edited excerpts from our conversation:

ENTERPRISEAM: What was your career before Gaia, and what shaped how you think about AI?

BADR AL MALLUH: My career has always sat at the intersection of operations, technology, and business. I majored in industrial engineering and later completed an Executive MBA at MBSC. I worked across Mars, Pfizer, and Panda, mostly in planning, supply chain, and project management. That experience taught me something important: companies rarely have a shortage of data. Instead, they have a shortage of accessible, contextualized information at the moment a decision actually needs to be made.

Later, working in startups gave me a much closer view of technology and product development, and I became increasingly interested in AI because I saw its potential to transform not just how people search for information, but how organizations actually operate. That became a key part of how I think about Gaia today: AI should not sit on the side of the business. It should become part of the operating layer.

E: Where did the idea for Gaia come from?

BM: We kept seeing the same problems and then went and validated them. We spoke with Saudi enterprise leaders across industries and heard the same things again and again: institutional knowledge getting lost, operational inefficiency, and global AI platforms that didn't meet Saudi regulations. That led us to a simple question: What if AI could understand the context of an entire organization while still operating within its security and regulatory boundaries?

E: What was the biggest challenge when you started, and how did you get past it?

BM: The biggest challenge wasn’t building another AI interface — it was making AI genuinely useful inside an enterprise where you have different systems, permissions, data formats, legacy infrastructure, security requirements, and regulatory constraints all having to work together. So we started with the enterprise architecture rather than the interface. Gaia is being built as an intelligence layer that connects to existing systems instead of asking companies to replace them, and we made security and sovereignty foundational requirements rather than features to bolt on later.

E: What are Gaia’s priorities from here?

BM: For now, the priority is execution — improve the product, work closely with clients, prove measurable value, and turn those deployments into long-term relationships. We’re starting in Saudi Arabia because it’s one of the strongest environments in the region for enterprise AI adoption.

Longer term, we want Gaia to become the trusted AI operational layer for organizations across the Kingdom and, eventually, the wider region, connecting people, knowledge, systems, and agents in one secure environment.

E: What’s the biggest problem you see Saudi businesses struggling with? And how does Gaia address that?

BM: Fragmented organizational knowledge. Companies have invested heavily in systems, but the information an employee needs is still spread across dozens of places. Finding an answer means knowing which system to look in, having the right access, searching through documents, then validating whether the information is even relevant. AI changes that equation because it can sit across those systems, understand the context, and actually execute actions using AI agents. But for a Saudi enterprise, that only works if it’s done securely, with the right data residency, compliance, permissions, and Arabic capability. That combination is what convinced us there was real potential, and it’s why we raised our pre-seed USD 1.5 mn.

Search is only the first layer. Gaia is designed to understand company context and give employees answers grounded in their organization’s own data, with sources — and from there, agents can interact with connected systems and run workflows. That’s the difference between enterprise search and an operational intelligence layer. The goal isn’t for someone to find a document faster — it’s to help them actually get work done across the systems they already use.

The strongest demand is exactly where employees lose the most time — searching, consolidating, validating, and moving information between systems — across functions like commercial, HR, operations, logistics, procurement, and engineering. What matters to us is that every engagement starts with a defined problem and a clear business case. We don’t want AI pilots that exist just to prove AI works.

E: What’s the agent / human split in tasks with Gaia?

BM: The AI agents handle repetitive, structured workflows across connected systems, while people stay in control of anything that carries real business, financial, regulatory, or operational consequences. For example, an agent can gather information from multiple systems, prepare an output, update a system, or kick off a workflow — but something like sending an email to an external client or a performance report to leadership can stay subject to human approval. The objective isn’t to remove people from the loop — it’s to remove unnecessary work from it.

E: Why would a Saudi business choose Gaia over an established global platform?

BM: Our advantage is that we’re building Gaia specifically around the realities of Saudi enterprises. That means sovereignty by design, local deployment options, a bilingual Arabic and English architecture, enterprise-grade integrations even with in-house systems, context-driven answers that understand how the organization actually works, and alignment with Saudi regulatory requirements.

What his mornings look like

I try to keep my mornings simple. I start by looking at what actually needs my attention rather than getting pulled into every message and notification, then identify the few things that can materially move the business forward that day and work through them with the team. As a founder, there’s always more to do than there are hours for it, so I’ve become increasingly focused on prioritization and speed with discipline — knowing what deserves attention and moving quickly once the decision is made.

Sports have played an important role in keeping the right balance in my life. Building a company can be intense, and sports give me a way to step away from work, reset, and come back with a clearer mind. They also teach you things that carry over into entrepreneurship, like discipline, consistency, and understanding that progress comes from showing up repeatedly over time.

I don’t think being a founder means focusing on work every minute of the day. You need the space to recharge so you can perform when it matters and not burn yourself out.

Al Malluh’s recommendations

On reading and learning: I don’t think there’s one book that’s had the biggest influence on me. I actually don’t read many books. I prefer staying curious and educating myself through online content, articles, research, and following developments in tech and business. Especially in AI, things move so fast that continuous learning matters more to me than relying on any one book. I try to make learning a habit and constantly expose myself to new ideas, perspectives, and developments.

A piece of advice that stuck: Start with a painful, measurable problem, not a technology. It’s very easy, especially with AI, to get excited about what the technology can do and then go looking for somewhere to apply it. Start with the problem instead. Is it painful enough that someone will pay to solve it? Can you measure the impact? Can you build something meaningfully better than the existing alternative? And once you find that problem, lock in on it. Don’t try to build everything at once.