Published: July 31, 2025 — This week, Brussels responded swiftly and strategically in the wake of Meta’s high-profile refusal to sign the European Union’s voluntary AI Code of Practice. With enforcement of the AI Act due in less than two weeks—on August 2—the EU is moving from diplomacy to action.

1. Meta Sparks a Showdown

On July 18, Meta Platforms became the first major U.S. tech giant to publicly reject the European Commission’s voluntary code of practice, labeling it legally vague and overly burdensome. Joel Kaplan, Meta’s Chief Global Affairs Officer, tweeted: “This code introduces a number of legal uncertainties … and goes far beyond the scope of the AI Act.”

2. Microsoft Takes a Different Path

In contrast, Microsoft indicated it would sign the code—welcoming industry engagement and regulatory clarity. President Brad Smith stated Microsoft’s goal is to balance innovation with compliance.

3. EU: From Guidance to Mandatory Regulation

The voluntary code, drafted by 13 experts, was meant to help companies document model training data, manage copyright risks, and improve transparency. Céline Regnier of the EU AI Office emphasized that non-signatories—like Meta—“will face closer scrutiny” once the AI Act is enforced.

4. Delay, But No Retreat

Although implementation of the code has been pushed to late 2025, the broader AI Act remains on schedule. Enforcement of general-purpose AI rules begins on August 2, 2025, with full obligations for existing AI models set to come into force by August 2027.

5. Momentum Camps Form

Meta’s move united a chorus of critics including Google, Airbus, and ASML, all citing innovation concerns. But other heavyweights—OpenAI, Mistral, and now Microsoft—have aligned with EU efforts. The standoff reveals two strategic tracks: one defined by voluntary engagement, the other by principled defiance.

6. Implications for Tech Giants and Startups

• Market Access Risks: Companies that don’t align with EU frameworks may encounter regulatory barriers in Europe.
• Legal Certainty vs. Flexibility: Signatories gain compliance protection; non-signatories will have to prove compliance individually.
• Innovation Geography: The growing divide may shift AI development toward more regulation-friendly regions.

7. Recommendations for Industry and Policymakers

• Track AI Act Implementation: Companies must monitor the August 2 rollout and subsequent enforcement phases.
• Conduct Transparency Audits: Even non-signatories should document AI training, data sourcing, and alignment policies.
• Engage in EU Dialogue: Join industry forums or consult with the EU AI Office to shape future regulations.

Bottom Line

The week of July 29–31 crystallized an emerging reality: voluntary benefits have prompted some AI firms to cooperate, while others have challenged the EU—sharpening a global split in regulatory strategy. With enforcement less than 48 hours away, the question isn't whether the EU will enforce, but how far it will go in enforcing—and how the tech industry will respond.

At WhatIsAINow.com, we’re watching these regulatory tectonics unfold so you can stay ahead of the storm.