Introduction
With sweeping AI regulations poised to reshape how artificial intelligence is developed and deployed in the EU, not all tech giants are on board. In a significant development, Meta has refused to sign the European Union’s Code of Practice for the upcoming AI Act—a move that signals resistance from global players ahead of a major regulatory shift.
But what does Meta’s refusal really mean for the AI industry, businesses building on AI models, and regulatory compliance moving forward? Let’s break it down.
What Is the EU’s AI Act?
The EU AI Act is the first comprehensive legal framework designed to regulate artificial intelligence across the European Union. It adopts a risk-based approach, classifying AI systems into categories such as:
Unacceptable risk: Banned uses like cognitive manipulation and social scoring
High risk: Applications in sensitive sectors like biometrics, education, and employment
Limited or minimal risk: Includes chatbots and recommendation engines
The aim is to ensure AI systems are safe, ethical, and aligned with European values—an ambition shared by many, though the methods remain hotly debated.
Meta’s Stance: Why It Refuses to Sign
In a post on LinkedIn, Joel Kaplan, Meta’s Chief Global Affairs Officer, voiced strong objections:
“Europe is heading down the wrong path on AI… Meta won’t be signing it.”
Meta argues that the Code of Practice for General-Purpose AI (GPAI) models introduces significant legal uncertainties and extends beyond the intended scope of the AI Act. According to Kaplan, this regulatory overreach risks stifling innovation and limiting the ability of developers to operate competitively within Europe.
Key Provisions in the EU’s Code of Practice
The voluntary Code of Practice, published earlier this month, is designed to prepare companies for the full implementation of the AI Act. Notable requirements include:
Transparency and documentation: Regular updates and disclosures about AI systems
Content usage restrictions: Prohibition against training AI on pirated or unauthorized content
Rights for content owners: Compliance with requests to remove proprietary content from datasets
Although voluntary, the code is seen as a precursor to mandatory rules, pressuring companies to align early.
Implications for AI Developers and Businesses
For developers and businesses, especially startups and mid-sized firms relying on general-purpose AI models, the stakes are high. Compliance involves complex documentation, risk management, and technical adjustments to align with legal expectations.
Companies must also assess:
Whether their models pose “systemic risks”
How to mitigate liability from third-party data use
The scalability of compliance across borders
For many, this could mean reshaping entire product pipelines.
The Wider Industry Reaction
Meta isn’t alone. Other tech giants—including Alphabet, Microsoft, OpenAI, Mistral AI, and Anthropic—have voiced concerns. In fact, several have lobbied the EU Commission to delay the rollout, warning that hasty regulation could put Europe at a competitive disadvantage.
Nevertheless, the European Commission remains firm. The finalized rules will apply from August 2, 2025, with a full compliance deadline of August 2, 2027 for models currently in use.
Read the official EU guidelines here.
Trenzest’s Take: Navigating AI Compliance Strategically
At Trenzest, we believe AI regulation—while challenging—can be a strategic opportunity. By aligning early with ethical and legal standards, businesses can position themselves as trustworthy AI providers.
What Comes Next?
With the deadline for compliance fast approaching, businesses and developers must begin internal alignment now. Those offering general-purpose AI models or integrating third-party systems should:
Conduct a gap analysis
Engage in legal and technical due diligence
Develop a roadmap for documentation and transparency
Ignoring the upcoming regulation could mean hefty fines and market exclusion, especially for those operating within the EU or serving EU customers.
Conclusion: Prepare for AI Regulation Today
Meta’s refusal to sign the EU’s Code of Practice isn’t just headline news—it’s a wake-up call. The debate between innovation and regulation is far from settled, but businesses must adapt to the new reality.
Whether you’re building AI systems or integrating them into your workflows, the message is clear: now is the time to prepare.
