Skip to content

AI code without copyright: licensing wash dispute threatens regional software

Share
AI code without copyright: licensing wash dispute threatens regional software

Listen to this article

Read by Anchor

Development teams in the Middle East are placing an accelerated bet on code-generation tools to speed up production, but the escalating dispute over the intellectual property of that software threatens to open unforeseen legal gaps for startups and tech firms in the Gulf, Egypt and the Levant. We followed this issue in the editorial office; it goes beyond mere speed of delivery, as the code produced by models without direct human intervention faces decisive claims that it must be made available as open source, and that it be deprived of copyright protection that regional companies rely on to safeguard their digital assets.

The first argument fueling this stance rests on training models on massive repositories of open-source code that have accumulated since Richard Stallman announced the GNU project and the free-software movement. That software was collected from the internet without permission or proper attribution to the original developers, and the models were also trained on content from Stack Overflow and its sister sites, which are subject to Creative Commons licenses that legally require source attribution. Turning this shared cumulative knowledge into closed, commercially owned code is described by observers as a licence-washing process that strips the programming community of its legitimate rights.

The legal dimension adds deeper complexity, because code generated entirely by artificial intelligence is not written by a human, which means it lacks the eligibility for copyright protection. This principle in U.S. courts is based on the well-known precedent of the “monkey selfie” case, which clearly held that non-human entities do not own intellectual-property rights, a precedent later relied on by the U.S. Copyright Office to reject protection for AI outputs. Similarly, the Wikimedia Commons platform is required to classify fully machine-generated images as public domain. Nevertheless, laws vary internationally; the United Kingdom grants limited protection to computer-generated works, creating a regulatory duality across borders.

The technical landscape is slipping in everyday practice; although coding is neither fully automated nor fully manual, developers have begun to rely on models to produce complete code without reviewing or adapting it to their specific needs.Code whose developer effort is limited to writing textual prompts loses automatic protectionwhich drives technical voices to call for publishing it under open licences or depositing it in the public domain via a zero-value Creative Commons licence, to avoid monopolising material not created by human thought.

What this means for the region

This dispute places a direct responsibility on technology leaders and company founders in Riyadh, Dubai and Cairo; building closed software platforms that depend on AI-generated code exposes the companies’ market value to accountability. If these codes are deemed to lack ownership rights, firms lose legal protection against competitors, and financing rounds or acquisition deals may stumble when technical due diligence examines the ownership of registered digital assets.

Exporting digital solutions from Egypt and the Levant or the global expansion of Gulf firms will clash with the differing legal environments between the United Kingdom’s limited protection and the United States’ strict refusal to recognise automated copyright. The alleged licence-washing also puts regional institutions at risk of compliance lawsuits over Creative Commons licences and the open-source code on which the original models were trained.

If you lead a technical team, the practical decision starts with configuring the workflow:Writing textual prompts alone does not grant your product intellectual-property immunity. Review your project's code, enforce a strict audit trail that demonstrates human intervention and architectural effort in modifying each piece of code, or adopt an open-disclosure policy for automated components to avoid upcoming legal disputes.

Don't miss the next story

Subscribe for updates