Saudi Arabia's National Strategy for Data & AI: what it means for your operations

The Saudi Data & AI Authority developed the National Strategy for Data & AI, approved by King Salman bin Abdulaziz Al Saud on 17 July 2020, under a vision it states plainly: "Where the best of Data & AI is made reality". It is not a technology roadmap for government alone. It sets the terms on which every organisation operating in Saudi Arabia is now expected to adopt data and AI — and it is worth reading as a specification rather than as an announcement.

Three horizons, not one deadline

The strategy is sequenced across three stages rather than pointed at a single date, and knowing which stage the Kingdom is in tells you what is expected of a deployment today.

That last stage is the one to sit with. The 2030 ambition is not adoption — it is export. A Kingdom that intends to export data and AI capability is a Kingdom where merely running a pilot is already behind the curve.

Six pillars, each with a target

SDAIA states the strategy's ambition as positioning the Kingdom as "the global hub where the best of Data & AI is made reality", and elevating it into what it calls the elite league of data-driven economies. Each pillar carries a measurable target rather than a slogan.

SDAIA also puts a precise number on the dependency: of Vision 2030's 96 direct and indirect goals, 66 are related to data and AI. That is the figure worth internalising. In the Saudi context, AI adoption is not a discretionary efficiency project sitting beside the national agenda — it is the mechanism most of that agenda runs on.

The five priority sectors

The strategy names five sectors for focus: education, government, healthcare, energy and mobility. If you operate in one of them, your AI programme sits inside a declared national priority, and that is worth saying explicitly in your business case. If you operate outside them — retail, financial services, logistics, telecoms — the strategy still reaches you through procurement expectations, regulation, and the talent market, but you will be making the argument on commercial grounds rather than borrowing the national one.

The six pillars, read as requirements

Below is what each pillar asks of an organisation actually deploying something, rather than of the country in the abstract.

Ambition

The bar set nationally is global leadership, not local adequacy. In practice that means a deployment which merely deflects calls is under-scoped. Pick use cases where the outcome is measurable in the business — resolution rate, revenue recovered, wait times eliminated — because those are the numbers that add up to a national position.

Skills

The pillar is framed as transforming the workforce with "a steady local supply of Data & AI-empowered talents". A strategy targeting more than twenty thousand specialists is telling you that capability is expected to reside in-Kingdom, in your team. Any AI deployment that leaves no local capability behind is at odds with that. Insist that your vendor transfers knowledge — that your own people can read the call reviews, change the escalation policy, and interpret the metrics without a support ticket.

Policies and regulations

The stated aim here is to "enact the most welcoming legislation for Data & AI businesses and talents", measured by a top-10 place in open data. SDAIA is both the strategy's owner and the Kingdom's data regulator, which is why the Personal Data Protection Law and the strategy should be read together rather than as separate obligations. Practically: know where personal data is processed, how long each type is retained, what a deletion request triggers, and whether callers are told they are speaking to an automated system. These are not blockers to adoption — they are the terms of it.

Investment

With capital deliberately flowing into the sector, the scarce resource is not funding but deployments that survive contact with production. Budget for the operating layer — weekly call review, prompt and policy change control, re-benchmarking when a model updates — not just for the build.

Research and innovation

Measured by a top-20 place in scientific contribution, this pillar is about producing evidence, not just consuming technology. Arabic is under-served by global AI relative to the number of people who speak it, and closing that gap is squarely in the national interest. Organisations contribute in an unglamorous way: by evaluating systems on real Saudi dialect data and telling vendors where they fail. Dialect performance improves when buyers measure it.

Ecosystem

A supported local supplier base is explicit policy. When you evaluate vendors, weigh whether they can be accountable in-Kingdom — reachable in your timezone, able to sit in a security review, able to put a named person on your weekly call review — alongside the global names.

What this changes about how you buy AI

• Scope for a measurable business outcome, not for a pilot that demos well.
• Treat PDPL alignment as an entry requirement and gather the data flow, retention and deletion answers before the commercial conversation.
• Require knowledge transfer so capability stays with your team, in line with the national skills ambition.
• Evaluate on Saudi dialect data from your own recorded calls — national ambition does not survive an agent that cannot understand a Riyadh caller.
• Budget for the operating layer, because AI systems drift and the strategy's targets are measured over years.

Where Voho sits in this

Voho builds Arabic-first voice operations for organisations in the Kingdom: Gulf, Najdi and Modern Standard coverage, PDPL-ready data handling, integration with the systems you already run, and an accountable operator reviewing real calls each week. We read the national strategy the same way we would read a customer requirement — as something you are measured against, not something you cite in a slide.