The Observation
Global publicly tried generative AI copyright cases reached 1,183 in early 2025. US federal dockets recorded a 327 percent increase in related filings.
The Analysis
The legal exposure of enterprise software is escalating rapidly. Companies deploy language models without verifying the origins of the training data. Employees generate marketing assets and code that violate intellectual property protections. The EU AI Act imposes fines up to 35 million euros for compliance failures. Organizations can no longer assume their vendor licenses grant blanket immunity from copyright infringement.
The Tactical Step
Conduct a strict data provenance audit for every generative tool in your enterprise stack. Restrict the generation of commercial assets until your legal team verifies the training data complies with regional regulations.
Question for the network
Has your legal team audited the training data of the AI tools your employees use daily?
By Michael Lennard Gnaedinger. © 2026 Gnaedinger Consultancy. All rights reserved.
If any of this sounds familiar.
I work with a small number of founders and CEOs each year. The conversation starts here.