ع
Global legal AI is consolidating around a handful of well-funded platforms. The regional question is whether the same logic applies in markets the incumbents were never built to serve.
The legal AI category is entering its reckoning year. After three years in which capital flooded the sector and hundreds of startups launched with overlapping promises, the market is beginning to sort itself, and the sorting is not gentle.
The capital picture tells the story. Legal-tech startups raised roughly $4.3 billion in 2025, some 54% above the previous year, reflecting the scale of capital moving into a category increasingly shaped by AI. But that money is not spreading evenly. Investors are writing the largest cheques to the platforms they believe will own the infrastructure layer, while placing smaller bets on specialist wedges where the category structure is still unresolved.
Some industry forecasters expect meaningful consolidation through 2026, with acquisitions, shutdowns and buyers narrowing around a smaller number of proven platforms. Sentiment among analysts has cooled in parallel: Forrester projects enterprises will defer a quarter of planned AI spending into 2027 over return concerns, with only 15% of AI decision-makers reporting EBITDA lift in the past year, while Gartner expects more than 40% of agentic AI projects to be cancelled by the end of 2027.
For anyone allocating capital or committing to a platform, the practical question is which companies survive that filter.
What the survivors appear to have in common
The emerging consensus is that capability alone will not be enough. Foundation models keep improving, and each generation absorbs more of what thin application layers used to sell. The platforms expected to endure are those embedded in systems of record and real workflows rather than those offering a prompt interface over someone else’s model.
Investors are reading the same signals. The strongest indicators, on this view, are repeat usage, proprietary legal context, integration into the systems where work actually happens, buyer trust, and measurable reduction of a legal bottleneck. Incumbents are moving accordingly. Thomson Reuters acquired legal-AI company Casetext for $650 million in 2023 and separately committed more than $100 million annually to AI capabilities, combining product development with acquisitions in its push toward the platform layer.
A regional field has begun forming against the same criteria, though it remains fragmented. Most of the platforms serving the market are built for a single jurisdiction, with Saudi-focused tools such as Adel and Shwra and Egypt’s LegalMind covering their home markets. Oqood AI, which covers Gulf jurisdictions rather than one of them, sits in the smaller group building across borders. That fragmentation is the structural fact the category has yet to resolve.
Which raises the question that matters for this region. If the global category is consolidating around platforms whose deepest datasets, product histories and customer bases developed in large English-language legal markets, what happens in markets with materially different legal and linguistic requirements?
The regional contest is a different contest
Consolidation logic assumes the consolidating platforms can serve the market they are absorbing. In the Gulf, that assumption meets structural facts the category has been slow to price. Across much of the region, authoritative legislation is published in Arabic, domestic legal systems are predominantly codified, and commercial practice routinely moves between Arabic and English.
The scale of the resulting gap can be measured. Arabic accounts for roughly 0.6% of content on the web, while Arabic speakers number approximately 400 million, close to 4.9% of the world’s population. A GCC Scope analysis puts Arabic at about eight times under-represented online relative to the people who use it, and in specialist legal material, scarcer and less structured than general text, the shortfall is steeper again. Models trained on the open web are not marginally disadvantaged in Arabic legal work. They are reasoning from a fraction of the material.
That is not a localisation problem that a well-capitalised incumbent solves by adding a language toggle. It is a different product, built on different data, validated against different legal authority. And it is why the funding pattern in this category matters less than it first appears: capital is consolidating around companies most likely to own the platform layer, but no comparable incumbent has yet established clear dominance over the Arabic legal-AI platform layer.
The regional signals point to a market forming rather than closing. Saudi Arabia designated 2026 its Year of Artificial Intelligence and is using the designation to accelerate national AI adoption and infrastructure. Ken Research values the GCC legal-technology market at around $1.2 billion, with Saudi Arabia and the UAE among its leading markets. Regional legal-tech adoption is climbing at roughly 15% a year, a rate that on GCC Scope’s calculation doubles adoption again in just under five years. And sovereign-linked capital is already positioned: Oqood AI, an Arabic-first legal platform covering Gulf jurisdictions, raised a $1 million seed round led by Sanabil, the PIF-owned investment company.
The size of the prize is easy to underestimate by looking only at law firms. The UAE’s Ministry of Justice automated services have processed over 125,000 transactions reaching more than 77,000 users. On a GCC Scope analysis, that single ministry has roughly ten times more users of its automated services than there are fee-earners in the entire GCC private bar. One ministry, in one of six states. The region’s roughly 900 law firms are the wedge into this market rather than its ceiling.
What to watch
Three things will determine how this category settles regionally.
The first is data. Building comprehensive, current and machine-readable collections of authoritative Arabic legal material is itself part of the product challenge. The shortage is documented: researchers behind the 2025 ALARB benchmark had to construct their evaluation from more than 13,000 Saudi commercial court cases because no substantial Arabic dataset for multistep legal reasoning existed to draw on. In a year when the category raised $4.3 billion, the most substantial public dataset for Arabic legal reasoning was assembled by academics out of court records. Whoever develops that layer gains an asset that is difficult to reproduce quickly, and it is precisely the kind of moat global investors are paying for elsewhere.
The second is institutional trust. In a market where professional bodies and regulators set the terms of practice, alignment with those institutions is a form of distribution that capital cannot shortcut. Oqood has moved earliest here, holding a strategic partnership with the Kuwait Bar Association alongside ISO and SOC 2 certification and a region-aware architecture built to meet different jurisdictions’ rules rather than impose a single foreign standard, as Continental Bulletin reported in July.
The third is enterprise readiness. As Gulf data-governance and cybersecurity expectations formalise, security certification, source traceability and configurable deployment stop being procurement details and start being the gate.
There is now early evidence of what clearing those three looks like in practice. Russell Bedford AlBaker, a Kuwaiti practice whose clients include sovereign-linked investment entities and family offices, set out figures to Justice GCC in August 2026 after integrating Oqood into live restructuring and regulatory mandates: first-pass bilingual contract review down from three to four hours to under an hour per document, preliminary statutory research from half a day to between 30 and 60 minutes, and eight to 12 hours a week recovered per fee-earner. The firm described these as internal estimates rather than audited benchmarks. Extended across a working year, a GCC Scope analysis puts that range at 368 to 552 hours per lawyer, or between nine and fourteen additional working weeks a year, with both core tasks compressing by a similar margin of roughly three quarters.
The global category is beginning to answer the question of who owns the platform layer. In MENA, that contest remains open, and the next several months start to test it.
Saudi Arabia’s Year of Artificial Intelligence runs to the end of 2026, with adoption and infrastructure outcomes to be measured against it. Forrester’s deferred enterprise spending is due to return in 2027, and Gartner’s cancellation forecast resolves at the end of that year. Closer in, the Kuwait Corporate Legal Summit on 21 and 22 October brings the region’s general counsel and managing partners together on exactly these questions, with Oqood founder Khaled Al Rashed among the panellists on legal AI in practice.
For investors and buyers, the gap between a consolidating global market and a regional category without a clearly dominant Arabic-first platform is the opportunity. How long it stays open, and who moves to close it, is what the coming year will show.
