Lexsoft T3 Becomes AI 'Knowledge Engine' with New Integrations

LONDON and MADRID – May 18, 2026 – Lexsoft Systems has significantly upgraded its legal knowledge management system, Lexsoft T3, positioning it as a central pillar in the rapidly evolving legal AI ecosystem. The company announced T3 is now fully integrable via the Model Context Protocol (MCP), and has simultaneously launched a new OpenAI-powered vectorized indexer for advanced semantic search. These moves transform T3 from a standalone repository into a dynamic, verified knowledge source for major AI platforms, promising to make AI outputs in the legal sector more accurate, reliable, and secure.

The 'USB-C for AI' Unlocks Legal Knowledge

The centerpiece of Lexsoft's announcement is the adoption of the Model Context Protocol (MCP). Introduced in late 2024 and now managed by the Linux Foundation's Agentic AI Foundation, MCP is rapidly becoming the industry's default standard for AI interoperability. Often described as the "USB-C for AI," the open-source protocol provides a universal interface that allows diverse AI models and external data systems to communicate seamlessly, eliminating the need for complex, custom-built integrations for each connection.

For the legal sector, this is a game-changer. By making Lexsoft T3 compatible with MCP, law firms and corporate legal departments can now connect their curated knowledge base to any MCP-compliant AI orchestrator, including prominent platforms like Microsoft Copilot, Claude, Gemini, and legal-specific AI like Harvey. This plug-and-play capability is a critical step forward. Previously, knowledge management systems often existed in silos, separate from the workflows where lawyers were beginning to use generative AI. Now, T3 acts as a "governed source" of truth, feeding high-quality, human-validated documents directly into the AI tools lawyers are using.

This integration addresses a fundamental challenge for AI in law: the quality of the underlying data. AI models are only as good as the information they are trained on. By drawing from T3—a system designed for the curation and lifecycle management of legal documents—AI orchestrators can produce outputs based on a firm's best, most current knowledge, rather than sifting through noisy, unvetted documents scattered across a general document management system. For every piece of information the AI extracts, T3 provides the exact contextual reference from the source document, ensuring full auditability and verifiability.

Beyond Keywords: A New Era for Legal Research

Concurrent with the MCP integration, Lexsoft has introduced a powerful new OpenAI vectorized Indexer. This technology fundamentally overhauls the search function within T3, moving beyond traditional keyword matching into the realm of true semantic search.

A standard indexer finds documents by matching exact words or their derivatives, such as "contract," "contracts," and "contracting." While useful, this method can miss relevant documents that use different terminology or return irrelevant results that simply share a keyword. The new vectorized Indexer, however, understands the conceptual meaning and context behind the words. It recognizes, for example, that "contract" and "agreement" are semantically similar concepts. It can also distinguish between different meanings of the same word, such as differentiating between 'Milan' the city and 'Milan' a person's name, based on the context of the document.

This is achieved by converting text into numerical representations called vectors. The system then finds documents not by matching words, but by identifying which documents are conceptually closest in this high-dimensional vector space. For legal professionals, this means more relevant and reliable search results. It enhances recall by finding conceptually related documents that a keyword search would miss, and it improves accuracy by filtering out documents that are not contextually relevant. This shift promises to dramatically reduce research time and surface insights that were previously difficult to find.

Fortifying Data Integrity in the AI Age

While the potential of AI is immense, law firms and legal departments remain acutely sensitive to issues of data security, client confidentiality, and regulatory compliance. Lexsoft's announcement directly addresses these concerns by architecting its new search technology with security at its core.

The new T3 Indexer is hosted inside the organization's own dedicated OpenAI tenant within their Microsoft environment. This is a critical distinction. It ensures that a firm's sensitive data and the indexes created from it remain within their own secure, controlled cloud infrastructure. This approach allows firms to maintain full control over data residency, helping them comply with regulations like GDPR, and benefit from the robust, enterprise-grade security architecture of Microsoft Azure.

Furthermore, because the data is processed within the client's private tenant, it is not used to train public-facing AI models, safeguarding client confidentiality and protecting proprietary firm knowledge. This separation of concerns provides a defensible framework for using powerful AI search capabilities on sensitive legal information. With Lexsoft T3 already holding ISO 27001 and ISO 27017 certifications, this new architecture further strengthens its position as a solution that balances innovation with the stringent security and governance requirements of the legal profession. Organizations are given a choice: they can continue using their existing Indexer or upgrade to the new, more powerful semantic search capability.

Powering Smarter, Safer Legal AI Workflows

By combining MCP interoperability with advanced semantic search and a human-centric review process, Lexsoft T3 is now positioned to play a vital role in the ongoing transformation of the legal industry. The integration turns the knowledge management system into an invisible but essential engine that powers a new generation of AI-driven legal workflows.

Carlos García-Egocheaga, CEO of Lexsoft Systems, emphasized this strategic shift. "By combining MCP functionality, advanced semantic search, and human‑centered review, T3 now plays a critical role in supporting the current wave of AI-led transformation of working practices in the legal sector," he stated. "We are developing T3 so that the solution enables organizations to optimally use AI for knowledge management while ensuring safeguards through close human involvement and oversight."

The ultimate goal is to embed high-quality knowledge management so seamlessly into the daily work of lawyers that it becomes an invisible part of the process. "The best part is that through generative AI tools such as Copilot, Claude, and Gemini, or Harvey; T3, knowledge management is seamlessly embedded within the legal workflow. In effect, the solution becomes invisible," García-Egocheaga added. This vision allows lawyers to trust that they are accessing the best, most relevant documents and precedents as a matter of course, enabling them to leverage the power of AI with confidence and efficiency.