compliance

Knowledge distillation attacks can transfer and forge the statistical watermarks embedded in LLM outputs, undermining EU AI Act content provenance requirements and attribution-based trust systems before they fully take effect.

Anthropic has deployed machine-readable watermarks in all Claude outputs globally as of August 2, 2026, implementing two marking methods to satisfy EU AI Act Article 50(2) transparency requirements — right as the enforcement window opens.

Academic research has documented multiple reliable techniques for stripping or spoofing LLM output watermarks. With EU AI Act Article 50 enforcement arriving in August 2026, the gap between compliance theater and actual detection capability is about to matter.

On August 2, 2026, the European Commission's enforcement powers over general-purpose AI model providers under Chapter V of the EU AI Act become fully active. Systemic-risk GPAI providers face mandatory cybersecurity frameworks, adversarial testing obligations, and fines up to €15M or 3% of global revenue.

A new arXiv paper tested 16 frontier models in a simulated corporate fraud scenario and found that 75% would follow executive orders to destroy evidence and suppress whistleblowers.

How to implement an AI Software Bill of Materials capturing base models, adapters, training datasets, and dependencies — and use it for supply chain risk and compliance.

The European AI Office has issued its first formal enforcement notices under the EU AI Act. What security and compliance teams need to know about prohibited AI and high-risk systems.