The market that changed shape overnight

In January 2026 the Saudi government declared 2026 the Kingdom's official Year of AI. What sounded like a slogan turned out to be a serious operating mandate: billions of dollars in AI-designated infrastructure, ministry-level AI KPIs, dedicated tenders for AI transformation across health, education, energy, finance, and government services.

Sovereign AI initiatives — Aramco Digital, the newly-launched Humain, G42's cross-border partnerships — are creating a market that didn't meaningfully exist two years ago. Agencies from the US, UK, India, and China are all trying to establish beachheads. Some are succeeding. Most are learning expensive lessons.

We've had a small team working on KSA engagements for eighteen months. This is the honest field guide — what actually works, what doesn't, and what to expect. It's the article I would have wanted before we started.

The three pillars of KSA's 2026 AI push

To understand the opportunity, understand the three flagship programs.

Aramco Digital

Saudi Aramco's digital arm, which now operates as a genuinely serious enterprise-tech buyer. Their mandate: modernize the operational stack of the world's most valuable company, and productize the outcomes for other regional customers. AI is central. They are hiring, contracting, and partnering aggressively.

For agencies, Aramco Digital is unusually pragmatic. They want proven capability, not slide decks. They pay well, on time, and expect delivery. Most engagements are multi-quarter, not multi-year.

Humain

Announced in mid-2025 as the Kingdom's national AI champion. Backed by the Public Investment Fund (PIF) with tens of billions committed. Building sovereign infrastructure — data centers, foundation models, applied AI products. They intend to be Saudi Arabia's Anthropic, Google Cloud, and Palantir in one entity, with international ambitions.

Humain is younger and more idealistic than Aramco Digital. They pay somewhat less but move faster, and they're building for the long term. For agencies, Humain engagements often involve co-development — you're not just delivering work, you're helping stand up their internal capability.

PIF-linked initiatives beyond these two

NEOM (with its recent scale-back, but still real budget). Red Sea Global. SDAIA (Saudi Data and AI Authority) with public-sector coordination. Vision 2030 program offices in almost every ministry now have AI mandates.

Each of these is a distinct buyer with distinct procurement rules. The mistake foreign agencies make is treating "Saudi Arabia" as one market. It's a portfolio of related markets, each with its own dynamics.

What KSA is actually buying

The high-value engagements we see falling into five buckets.

1. Sovereign LLMs and Arabic-first AI infrastructure. Humain and partners are training regional LLMs with strong Arabic and Islamic cultural competence. Agencies with model training, data engineering, or eval expertise are in demand.

2. AI transformation for large state-owned enterprises. Aramco, SABIC, STC, Saudi Airlines. These are mature companies with real budgets going through genuine agent-led transformation. Change management, integration engineering, custom AI applications.

3. Public-service AI. Health system agents. Education tutoring platforms. Ministry-facing chat interfaces for citizens. Immigration and Hajj/Umrah services. These are big-ticket, high-visibility, politically sensitive.

4. Financial services AI. The Saudi Central Bank (SAMA) has been aggressive about AI in banking. Agentic customer support, fraud detection, credit decisioning — all live investments.

5. Content and creative AI. With the Kingdom's massive tourism, entertainment, and sports investments, content generation, dubbing, translation, and cultural adaptation are surprisingly big line items.

If your agency's capability doesn't fit one of these buckets, forcing your way in is expensive. Better to lead with what genuinely fits.

The barriers foreign agencies routinely underestimate

Let's be blunt about what's hard.

Relationships are the market

KSA operates on trust and personal introductions to a degree that surprises Western agencies. Cold outreach fails. RFP responses without a prior relationship rarely win. What works: sustained presence, real partnerships with local firms, senior team members in-country regularly.

Agencies that come in for a quarter, land a deal, then disappear are read as "not serious." Agencies that stay through the slow build are the ones that eventually win big.

The procurement pace is bimodal

Government and semi-government procurement in KSA can take 12-18 months from RFP to contract signature — longer than most Western markets. But once a program is prioritized and top-down direction is given, decisions get made in days. Your job is to be positioned so that when the fast phase happens, you're the obvious choice.

Local partnership is often required

Many KSA contracts require partnership with a Saudi entity — either a local firm or a Kingdom-registered subsidiary. Building this correctly takes months. Don't wait until you have a hot lead; establish it before you need it.

Cultural and religious literacy matters

This isn't a soft skill. This is table stakes. An AI product that stumbles on Arabic diacritics, misidentifies prayer times, or produces content inconsistent with Islamic values will not be renewed. Right-to-left support is not "an internationalization ticket." It's a core design constraint.

For AI systems specifically: your models need to handle Arabic natively (see our small language models piece — Qwen and Falcon are better here than most defaults). Your evals need to include Islamic cultural competence. Your training data needs to be reviewed for cultural fit.

Data residency is non-negotiable

Personal data of Saudi residents, and often the outputs of AI processing that data, cannot leave the Kingdom for many use cases. This is enforced. Cloud infrastructure choices need to match — AWS Middle East (Bahrain), regional Azure, local Aramco or STC cloud partnerships. Vendors without a compliant deployment path get filtered out early in procurement.

What we've learned actually works

After eighteen months of engagement, here's the pattern that generates real business.

1. Invest in relationships before you have a pipeline. Attend LEAP, Global AI Summit Riyadh, Money 20/20 Middle East. Be in Riyadh multiple times per year. Get introduced.

2. Partner with a local firm or Kingdom-registered entity. Not because it's required for every deal, but because it makes every deal easier. A local partner unlocks access, provides cultural translation, and shares risk.

3. Lead with a small, deliverable engagement. A 6-8 week strategic AI assessment or a well-scoped pilot. Prove capability on something real before pitching a transformation program.

4. Deploy senior people, not junior consultants. KSA buyers expect gravitas. Sending a junior engineer as the primary contact reads as disrespectful. Match seniority to the client's expectations.

5. Ship Arabic-first, not Arabic-later. If your product will be used by Arabic speakers, design for Arabic from day one. Retrofitting always shows.

6. Understand Islamic finance if you're in fintech. Sharia compliance, tawarruq, murabaha — these are real product constraints for financial-services AI. Ignorance is not a good look.

7. Be patient about the enterprise sales cycle. Six to twelve months from first meeting to contract is normal for mid-sized engagements. Anchor internally on that.

The Bengaluru-to-Riyadh angle

For Indian agencies specifically — which is where we're based — there's a meaningful advantage worth noting.

• The talent story resonates. India's technical depth is respected, and cost-competitive relative to Western agencies.
• Cultural and religious empathy is easier for teams with Muslim members and Arabic-speakers, and Bengaluru's talent pool includes many.
• Time-zone overlap is workable. Riyadh is 2.5 hours behind Bengaluru — real-time collaboration is comfortable.
• Direct flights are frequent and short.
• Historical trade and diaspora ties smooth the introduction process.

But the advantage only holds if you show up. Indian agencies that stay in India and bid remotely are treated the same as any distant vendor. The advantage compounds only if you invest in on-the-ground presence.

The Aramco Digital and Humain angle for AI-specialist agencies

If your specialization is AI (as ours is), a few specific patterns:

• Aramco Digital tenders often require demonstrated experience with industrial AI, agents that handle physical-world data, and integration with SCADA/OT systems. Pure SaaS AI experience is not sufficient.
• Humain engagements often require the ability to work alongside their in-house team as a capability transfer, not just a delivery vendor. Set expectations accordingly.
• Both will ask about your position on sovereign models — are you willing to build on Humain's models, or do you insist on frontier APIs? The right answer is usually "we'll build on either, and here's how we evaluate the trade-off."

Frequently asked questions

Do we need to open a Saudi legal entity? For serious pursuit, yes eventually. For initial engagement, a strong local partnership can substitute. The threshold depends on client type — MoU work sometimes fine without; direct government contracts often require it.

What's the average deal size? Highly variable. Pilots and strategic assessments: $80K-$300K. Mid-sized delivery engagements: $500K-$3M. Multi-year transformation programs: $5M-$50M. The top of the range requires established relationships and delivery track record.

What Islamic-culture considerations affect AI products specifically? Content generation must not produce material inconsistent with Islamic values. Financial products must have Sharia-compliant paths. Health AI must handle gender segregation appropriately in certain contexts. Location-aware apps must integrate prayer times and Ramadan schedules gracefully. These are product-design considerations, not afterthoughts.

Is English enough or do we need Arabic-native team members? English is workable for many senior interactions. Arabic-native team members dramatically improve depth of engagement, especially in government and traditional-industry contexts. If you don't have Arabic capacity, partner with a firm that does.

How does the Vision 2030 slowdown / NEOM scale-back affect the AI market? Less than the headlines suggest. Vision 2030 is being rescoped, not abandoned. NEOM's scale-back means less speculative future-city work but does not affect the core AI budgets in health, education, finance, and energy. Those remain strong.

Are Chinese vendors dominating? Some verticals, yes. In cloud infrastructure and hardware, Chinese vendors have significant footprint. In AI applications and consulting, the field is genuinely open — Western, Chinese, Indian, Emirati, and Saudi vendors all compete. Best product and best relationship wins.

The strategic read

Saudi Arabia's Year of AI is not marketing. It's a real, well-funded, politically-mandated push to make the Kingdom an AI-leading economy in five years. The market opportunity for agencies who show up, deliver, and stay is genuinely large. The barrier for agencies who treat it as a one-off is genuinely high.

If your firm is serious about the Middle East, 2026 is the year to establish position. Waiting until 2028 puts you behind competitors who used this year to build relationships.

At Xenolve we run engagements across the Gulf from our Bengaluru base, with regular presence in Riyadh, Dubai, and Manama. If your team is evaluating KSA as a strategic market and wants an honest partner conversation — including whether the market is right for your capability — get in touch. We're happy to compare notes even if we don't end up working together.

The Kingdom is open for AI business. The question is whether you're doing the work to be part of it.