A memorandum of understanding signed in Abu Dhabi targets the point where most corporate AI projects stall: the gap between the pilot and the production system.

Corporations have been pouring money into artificial intelligence, but the result has not mirrored the effort. A widely cited 2025 study from MIT’s NANDA initiative titled The GenAI Divide: State of AI in Business, found that roughly 95 percent of enterprise generative AI pilots delivered no measurable impact on profit or loss, even as companies spent an estimated $30 billion to $40 billion trying. The problem it described was not model quality or regulation. It was the unglamorous work of getting AI out of the demo and into the daily workflow.

That gap is the backdrop for a modest looking document signed in Abu Dhabi on July 13. Inception42, the AI developer that sits inside the G42 ecosystem, signed a memorandum of understanding with NXT Holding, an AI-native holding company owned by Sahm Holding. On paper it is a framework agreement covering five areas: AI consulting, training and upskilling, enterprise AI solutions, product distribution, and joint business development. Read against the MIT findings, it is something more specific - a bet on distribution.

A Channel Play, Not a New Model

What is notable about the agreement is what it does not contain. There is no new model, no chip, no data center. Instead, Inception42 built the technology positioning itself as the intelligence layer of what G42 calls its Intelligence Grid. NXT Holding brings something the AI industry has learned to value the hard way: reach. Its portfolio of operating companies gives Inception42 a set of real businesses to plug their products into, along with the investment muscle to fund those deployments and the local relationships to sell them.

The MIT research points to why that matters. The organizations that crossed the divide were not the ones with the best demos, they were the ones that treated AI as a partnership rather than a purchase, held vendors to business metrics, and did the integration work most companies skip. A distribution agreement is precisely the sort of plumbing that separates the 5 percent from the rest. The same research found that buying from specialized vendors and building partnerships succeeded roughly 67 percent of the time, while internal builds worked only about a third as often. That single contrast reframes a distribution deal as a strategy rather than a convenience.

Ashish Koshy, chief executive of Inception42, framed the deal in those terms.

“Inception is built to deliver sovereign, enterprise-grade AI at scale. Partnering with NXT Holding expands our ability to reach organizations that are ready to move from AI exploration to AI execution. This MoU enables us to combine our product and platform capabilities with NXT Holding’s market access and investment expertise, creating a powerful channel to drive measurable AI impact across the UAE and beyond.”

Why “Sovereign” is the Selling Point

Both sides lean heavily on one word: sovereign. In the Gulf that signals AI that runs on locally governed infrastructure, with data staying inside national jurisdiction rather than being handed wholesale to foreign clouds.

The timing tracks a national push where the UAE has set a target for AI agents to support half of federal government operations within two years and plans to train tens of thousands of public sector workers to use them, with Abu Dhabi aiming to become the world’s first AI-native government by 2027. The Inception42 and NXT Holding agreement extends the same logic outward, from federal ministries toward the wider private sector.

The Holding-Company Bet

NXT Holding represents a structure that is becoming more common as AI matures: the holding company built expressly to spread the technology across everything it owns. Rather than treating AI as a product line, Sahm Holding has organized an entire vehicle around the assumption that AI will reshape how its businesses operate, and that owning the channel into those businesses is itself an asset. It is a wager that in the AI era, distribution is worth as much as the technology being distributed.

Fahmi Abubakar, managing director of Sahm Holding, described the rationale plainly.

“NXT holding was built for the structural shift that artificial intelligence represents in how economies and institutions operate. Our responsibility is to help governments and enterprises navigate that shift. The agreement with Inception expands on our delivery commitment, and we look forward to the value it will create”.

The political weight behind the deal is hard to miss. Sheikha Fatma bint Tahnoon bin Zayed bin Sultan Al Nahyan, founder and chairwoman of Sahm Holding, tied it directly to the country’s strategy.

“The UAE has placed artificial intelligence at the center of its national vision, and partnership of this nature are how that vision becomes capability. This agreement reflects the strength of our institutions, the alignment of their purpose, and the nation’s determination to lead in technologies that will define the future.”

In a market where state ambition and private capital move in close coordination, that endorsement is a signal that the partnership is meant to serve national goals.

What Comes After the Signing

A memorandum of understanding is not a contract, it commits the two sides to explore, not to deliver. The framework touches five areas, but it names no timelines, no spending, and no first customer. Its value will be measured by whether AI actually reaches production inside NXT Holding’s companies and the outside clients it courts.

That is the appropriate lens for a deal like this. The ambition is real and the strategy is coherent, aimed squarely at the point where most enterprise AI dies. The proof will be operational, and it will arrive quietly, in systems that keep running long after the signing photos are forgotten.

Related Articles

Inception42 and Microsoft Launch Seraj, an Enterprise AI Model Built for Arabic Fluency

How 1001 Raised $30 Million to Bring Sovereign AI to Ports, Airports and Energy Grids

How Saudi Arabia’s IT Leaders Are Building the Infrastructure Layer for Enterprise AI