Some law firms are developing in‑house artificial intelligence tools instead of purchasing standard solutions, because off‑the‑shelf tools do not always adapt well to the specific needs of a practice, a jurisdiction or a document base. In some cases, these tools are also offered to clients on a subscription basis, opening the door to new revenue models.
The Financial Times describes this phenomenon as the rise of “build‑your‑own” within law firms.
The trend of “in‑house” AI software in law firms
The Financial Times describes a growing phenomenon: law firms are moving beyond the idea of purchasing “off‑the‑shelf” solutions to develop, internally or with technology partners, tools optimized for real workflows, precedents and client cases.
In several regions of the world, from Singapore to Sydney, major international firms are building proprietary generative and predictive AI, powered by their own databases and internal processes.
This shift in approach can be seen as the result of several specific factors: the maturity of generative AI technologies and the pressure on costs and operational efficiency. Added to this is a structural limitation of market solutions: they often do not adapt to the specificities of a jurisdiction, a sector or a particular client profile. In these cases, developing an internal tool can become the most suitable solution.
Real‑world cases
Among the examples cited are Allen & Gledhill and Rajah & Tann in Singapore, Inkling Legal Design in Sydney and Ashurst in London. These cases show a recurring model: AI does not replace legal oversight, but amplifies it.
The Singaporean firm Allen & Gledhill developed in 2024 A&GEL (“angel”), a generative AI tool designed to support practice groups. The tool accelerates high‑volume activities by comparing drafts with verified and constantly updated precedents, thus maintaining high reliability standards. The goal is not to reduce lawyers’ work, but to improve their operational quality without compromising the robustness of the analysis.
The value therefore also shifts to the quality of the output and the ability to adapt it to context.
Inkling Legal Design has also chosen to design its own tool. The firm’s predictive tool uses historical user‑testing data across multiple jurisdictions and flags potential issues such as overly complex language or cultural sensitivities. Sara Rayment, founder and managing partner, explains that existing AI products on the market were often trained on U.S. legal documents, which are more aggressive and protective from a contractual standpoint. In practice, they tended to propose clauses, formulations and structures typical of the U.S. context, which do not always fit well with clients or markets with a different legal culture. The issue was not only linguistic, but above all structural: those models were “weighted” to give more importance to recurring examples and logics found in American material. As a result, the output was less useful for Asian, Australian or British clients, because the way a contract is written, negotiated and interpreted can vary greatly from one legal system to another.
Here too, the reason behind the choice is adaptation: the legal market is not uniform, and generalist tools do not always meet its specific needs.
From internal software to client services
Operational transformation directly affects the economic model. If AI drastically reduces the time required for activities such as document review or drafting, the traditional billable‑hours model must be reconsidered. This is why hybrid models are emerging that combine technology and human expertise.
Rajah & Tann, in Singapore, has launched a subscription service that provides access to a contract‑review platform based on its internal systems. Rajesh Sreenivasan describes the model as a subscription for the “vault” plus an hourly fee when the support of the legal team is needed. The service is based on access to the firm’s digital repository, which contains verified legal documents and data. In this way, the client does not purchase only a technological tool, but enters an organized environment of controlled legal content, with the possibility of returning to the firm for more complex matters.
Thus, AI does not eliminate legal work, but helps shift the focus toward more strategic and complex activities.
Ashurst is rethinking its business model by mapping the activities that generative AI can replace and building alternative billing formulas that combine technology and human expertise. Hilary Goodier, global head of Ashurst Advance and of the firm’s digital transformation, explained that AI is placed at the center of the service, around which processes and people are built.
A sector still at the beginning
Goodier notes that the sector is still at the beginning of this journey and that the past two years have seen more technological progress than the previous twenty. This observation helps frame the phenomenon: we are not facing a change that has already been completed, but a transitional phase in which only some firms are experimenting with different models. The picture that emerges is that of a legal profession which, in some cases, not only uses AI but begins to build it, adapt it and, at times, distribute it to clients.
The case of the MIKE platform (OSS AI Legal Platform), an open‑source project for contract drafting available on GitHub, also moves in this direction: tools that are more adaptable and built around the specific needs of legal practice. In the Italian and European context, the theme of customization carries particular weight, because GDPR and professional liability require strict control over data, models and processes.
The value of data
One point on which the experiences analyzed converge is that competitive advantage does not lie only in the algorithm, but above all in the data. In this context, informational assets become the true engine of value: data quality, governance and organization matter as much as—if not more than—the technology itself.
Sources:
Marta Magnini
Digital Marketing & Communication Assistant at Aidia, graduated in Communication Sciences and passionate about performing arts.
