Writing

Abu Dhabi’s Chamber chose one vendor for 102,000 SMEs. Here is the independent alternative.

In June 2026 the Abu Dhabi Chamber of Commerce and Industry signed a partnership with Presight, a G42 company, to bring agentic AI to the more than 102,000 small and medium businesses registered with it. It is a serious programme from serious institutions. It is also one approach among two, and an owner deciding what to do about agentic AI should understand both.

What was announced

The Abu Dhabi Media Office and Presight describe the partnership in similar terms. Presight will deploy what it calls sovereign AI systems, built to deliver scalable and secure capabilities tailored to SME environments, across the Chamber’s network of more than 102,000 registered SMEs. The stated aim is to let those businesses embed operational intelligence into their processes: autonomous workflows, real-time decision support and continuous optimisation. The roll-out starts with a pilot cohort and follows a structured roadmap to scale across the wider base, and the Chamber will offer its members access at preferential rates.

There is nothing to quarrel with in any of that. A chamber’s job is to raise the floor for all of its members at once, and a single well-resourced platform, deployed by a company with the scale to support it, is the right instrument for that job. If the programme does what it describes, a very large number of businesses will have better tools than they had before, at a price they could not have negotiated alone. That is a good outcome and we hope it happens.

What a platform roll-out can do

It can reach everyone. A platform built once and deployed to a hundred thousand companies has a cost per company that no bespoke approach can match, and it arrives with the support, security and continuity that a large provider can promise. It can standardise. Businesses in the same sector doing the same things in the same way can be given the same workflows, and the aggregate improvement across a sector is real. It can be measured, because the same thing was deployed everywhere and the same numbers can be read back. And it can be done quickly, which matters when the agenda has a date on it.

For a business whose operation is close to a standard shape, these are decisive advantages. A trading company with a familiar order flow, a services firm whose jobs look like each other, a shop: a well-built platform will fit well enough, and the owner should take it.

What it cannot do

It cannot know how one particular business actually runs. That is not a criticism of any vendor; it is what the word platform means. A platform is built around a model of a business, and every company is then fitted to the model. For most companies the fit is acceptable. For an owner-run company that has grown to twenty or forty people on the strength of doing things its own way, the fit is where the trouble starts, because the things that make the business work are precisely the things the model does not have a field for.

We see this every time we sit in a room and map an operation end to end. The process a production company actually follows from the first call to the six-month follow-up, the way a coaching business really moves a lead from a booked call to a payout, the point at which a commerce brand loses the customer who never reorders: these are specific, they are the business, and they are not in any template. A platform can hold the parts that are standard. The owner still carries the joins, and the joins are where the operation lives.

There is a second limit that is structural rather than technical. A platform deployed to a hundred thousand companies has, by design, a hundred thousand customers, and the direction of the product is set by the aggregate rather than by any one of them. When a business needs the system to do something it does not do, the answer is a feature request. That is reasonable for a platform. It is a constraint for a company whose competitive edge is the way it operates.

The case for building one company at a time

The alternative is slower, narrower and more expensive per company, and it is the right answer for a specific kind of business: owner-run, with real operational complexity and real revenue, where the founder is the bottleneck and the operation is the asset. For that business the question is not which tools to adopt. It is how to get the operation out of one person’s head and into a system that runs it, and the system has to be shaped by the operation rather than the other way round.

That is what we do, and the method page describes it: a day in a room mapping one real job end to end, with every point where something is dropped on the wall; then one system, the four parts described under the operating system, built around what the map showed, on the company’s own infrastructure and in its own brand; then an executive on it, one senior AI that runs the system and escalates to a named person, with no authority of its own. Three such systems exist, two in daily use and one in build, and they are described under work in production.

The honest comparison is this. A platform roll-out is how a chamber raises the floor for everyone. A system built one company at a time is how a particular company gets a ceiling that is its own. The two are not in competition, and most of the 102,000 businesses in the Chamber’s network are better served by the first. A smaller number of them, the ones whose operation is their edge, are the companies we exist for.

How an owner should decide

Start with the operation, not the technology. If the way the business runs is close to how similar businesses run, a platform will fit, and the Chamber’s programme or one like it is the sensible place to begin. If the way the business runs is the reason it wins, and the owner is the only place the whole of it exists, a platform will hold the standard parts and leave the owner carrying the rest. The build or buy note sets out the questions to ask, and they apply to a chamber’s programme as much as to any vendor. Whichever way the answer falls, the decision is the owner’s, and the point of writing this is that it should be an informed one.

Written by the founders of Symbaiotic, Dubai. See who we are, and more writing.