Autonomous and intelligent documents

 

In an era of increasingly intelligent and autonomous cars, we also need to consider documents that behave as intelligently and autonomously as possible. Traditionally, documents lead a dull life. They sit unused and forgotten, and in the worst-case scenario – if, against expectations, they’re needed again after a long period of time, for example by Compliance – they sometimes can’t be found at all.

Gutenberg put a lot of effort into making the fonts used by his pivotal printing press look like handwriting. He didn’t aim to make reading easier for the user – that is, the reader. The same error is in evidence in today’s digitization projects: digitized documents are intended to look like printed paper. The need to carry out automated analyses of document content often goes unrecognized. We’re now going one conceptual step further and considering how we can create intelligent, autonomous documents.

Ideas for intelligent documents can be found in the automobile industry. Requirements include sensors, behavioral and control rules and actors that can carry out actions. Behavioral rules are instantly reminiscent of business rules. There’s nothing to prevent documents from containing automatically interpretable information in order to deploy rules regarding data protection or record retention obligations. The rules themselves do not belong to the documents, as they can change – for example when a legal hold is implemented.

Libraries’ traditional control slips can contribute another idea: these slips indicate the process that a newly acquired book has to go through. The slips are stuck between pages and provide information about a book’s processing stage at all times. The steps it has already gone through are fully traceable. The digitalization of a control slip requires the allocation of a unique identifier. Individual process steps can then be easily visualized. By linking the process steps documented in this way, traceability can be guaranteed. This information does not necessarily have to be organized in linked blocks, but it does suggest a conceptual relationship with blockchains, which are currently attracting a lot of attention. Blockchain characteristics such as privacy, traceability and independence from a central monitoring body offer important characteristics for document management in the future.

Data Protection Act

The Swiss Data Protection Act underwent a complete revision in 2020, and its new version took effect on September 1, 2023, along with the new Data Protection Ordinance (DPO). The revision itself is complete, but its practical application continues to evolve. The topic of AI is particularly relevant: On May 8, 2025, the FDPIC confirmed that the DPA is technology-neutral and applies directly to all AI applications. In practice, this means, among other things, that users must know whether they are interacting with AI (transparency requirement, Art. 19), and that a data protection impact assessment is mandatory in cases of high risk—such as profiling or facial recognition (Art. 22). The Federal Office of Justice (FOJ) is drafting a consultation document on AI regulation to be completed by the end of 2026. With this, Switzerland will implement the Council of Europe's AI Convention.

Eurospider Information Technology AG
Winterthurerstrasse 92
8006 Zürich

 

Cookies make it easier for us to provide you with our services. With the usage of our services you permit us to use cookies.
More information Ok Decline