Playing Soft with the Rules
A new report from Euronews reveals that major tech firms like Google, Microsoft, and Meta significantly diluted the EU’s voluntary AI Code of Practice, transforming it into a weaker framework. Originally intended to block harmful AI content such as deepfakes and disinformation, the final result sees vaguer language and ambiguous commitments. Experts claim industry lobbying shaped the outcome, stripping away enforceable obligations under pressure. Critics now say what was meant as a strong, self-regulating tool has become a toothless PR move.
Transparency Lost in Translation
Despite the EU’s initial intentions for more transparency in AI systems, companies reportedly pushed back against requirements that would demand clear labeling, stronger content moderation, and traceability mechanisms. Instead, they advocated for looser definitions and favors around “flexibility” and “innovation.” The resulting framework, while still in place, now relies heavily on self-assessment with minimal external oversight. This opens the door to potential misuse or non-compliance as AI systems scale rapidly across platforms.
Echoes of the AI Act Debate
The watering down of the code casts a looming shadow over the upcoming EU AI Act, which seeks to introduce legally binding guardrails. Observers warn that if voluntary frameworks can be derailed, future legislation may face similar pressures from tech giants wielding economic and political influence. With generative AI and misinformation posing growing risks to democratic processes, policymakers are under mounting pressure to ensure the next set of rules has real teeth.
