Digitalization Forum: Off-the-shelf Software Is No Longer Secure: Reverse Engineering via AI Also: New GDPR Exceptions for AI Training
Until now, the general assumption was that software object code is sufficiently secure. No one can meaningfully reverse-engineer the software based solely on the object code. This is currently changing. This is because AI can now be used to generate source code for any object code. As a result, replicas can be created, and competitors can gain insight into previously secret processes within your software. But is this even allowed? Learn more about this at the Digitalization Forum on November 25, 2026.
We’ll also examine the latest proposals related to the “Digital Omnibus” reform package, which provides for far-reaching exemptions from the GDPR when this serves the purpose of AI training. This could make data uses that were previously impossible now feasible. We’ll take a look at the current status of the draft legislation.
On the one hand, reverse engineering is gaining massive momentum through AI-powered analysis tools. For established developers of proprietary software, control systems, and medtech solutions, this poses a significant risk: modern AI makes it possible to reconstruct code, logic, and interfaces more efficiently than ever before. At the same time, the technology offers users—especially hospitals and device manufacturers—new ways to break free from costly, proprietary lock-in effects and achieve interoperability. This raises urgent questions: When are AI outputs protected? Where does legal reverse engineering under the German Copyright Act (UrhG) and the German Trade Secrets Act (GeschGehG) end? And how can one effectively safeguard one’s own development assets?
On the other hand, the Digital Omnibus—which is still in the European legislative process—heralds a far-reaching reform of data regulation. This legislative package aims to harmonize existing legal acts, reduce bureaucracy, and strengthen data-driven innovation. For research-based companies, medtech pioneers, and the healthcare sector, this presents enormous opportunities: Among other topics, the discussion will focus on regulatory relief for the legally compliant secondary use of sensitive health and treatment data for research, development, and AI training purposes.
As part of the Digitalization Forum on November 25, 2026, from 3:30 p.m. to 5:00 p.m., Dr. Gerrit Hötzel and Marius Adler will provide a concise, interdisciplinary, and well-founded overview of these two key areas of the future:
AI Outputs & Reverse Engineering in Practical Application: Copyright classification of AI results, limits of trade secret protection, threats to established software and system architectures, and strategies for overcoming lock-in effects. The speakers will present concrete practical solutions and strategic protection mechanisms to reliably safeguard know-how and code investments.
Outlook: Opportunities Presented by the Digital Omnibus: A strategic outlook on the reform package—with a particular focus on future rules governing data use, measures to facilitate data-driven innovation, and the emerging potential for training AI models in the healthcare sector and the medtech industry.
Take this opportunity to learn about the new legal dynamics of AI-driven software competition, gain practical protection measures for your systems, and stay ahead of the curve regarding future data protection regulations for your company or organization.
The event will take place online and is free of charge.