InvizCrypt, developed at Trinity College Dublin, is designed so users can collaborate on sensitive documents without the platform provider being able to read the content
As European governments and institutions move to reduce reliance on large US technology platforms, researchers at Trinity College Dublin have developed a new cloud collaboration platform designed to give users stronger control over who can access their sensitive data and documents. InvizCrypt, led by Professor Hitesh Tewari of the School of Computer Science and Statistics and ADAPT at Trinity College Dublin, is designed to allow users to collaborate on sensitive documents while preventing the platform provider from reading the content.
The project comes amid growing European concern about data privacy, security, digital sovereignty and foreign legal access to sensitive information held by large technology providers. In April 2026, the European Commission advanced a sovereign cloud procurement framework that will allow EU institutions, bodies and agencies to procure up to €180 million in sovereign cloud services over six years. France has also moved to replace Microsoft Teams and Zoom with its own sovereign video conferencing platform, Visio, across government departments by 2027.
Speaking about InvizCrypt, Professor Hitesh Tewari said: “Users place huge trust in cloud platforms to store and process their data but many people do not realise that, in conventional systems, the provider can technically still access document content. That creates risks around privacy, security and legal disclosure, regardless of where data is hosted. InvizCrypt is designed to offer a fully private model that can not only support real-time collaboration but also ensure the security of the data within the documents.”
Most cloud collaboration platforms are built on a model where the provider can technically access the contents of a document as part of delivering the service. The InvizCrypt system is designed differently. Before a document leaves the user’s own device, it is encrypted into unreadable data. The cloud platform can still store the file, allow collaboration between users, and keep the document in sync, but it cannot see what the document says. Only approved collaborators hold the keys needed to read it. This fundamental difference means that documents, attached files and real-time edits are protected before they reach the platform. For users, InvizCrypt feels like ordinary cloud collaboration but with the added layer of security.
The first users of InvizCrypt will be researchers because research teams often work on highly sensitive material before publication. This information can be valuable to competitors, cyber criminals or even foreign state-backed actors, and universities are increasingly being warned that research data and intellectual property are attractive targets. The team has started with LaTeX, a writing system used by many researchers, scientists, engineers and mathematicians to prepare complex academic papers and technical documents. Following this first phase, the technology will then be extended into secure document editing and spreadsheet workflows for organisations that need to collaborate without exposing sensitive content to platform providers.
By beginning with an encrypted collaborative LaTeX editor, InvizCrypt targets a community that already depends on cloud based collaboration but has limited options for keeping sensitive work private from the platform provider. This initial phase will allow the team to validate the system in real research workflows, build trust with universities and research institutions, and establish the institutional relationships needed for wider adoption.
The team will then extend the same server-blind architecture beyond LaTeX into secure document editing and spreadsheet workflows, creating a broader privacy-first platform for a range of users and domains later in 2026.
