Overview
Severity: MEDIUM | Affected: EU-based AI Developers | Category: policy
The European Parliament has passed the 'AI Liability and Auditing Directive' (AILAD), a landmark piece of legislation that builds upon the foundational EU AI Act. The new directive establishes a strict liability framework for damages caused by high-risk AI systems, making it easier for victims to seek compensation. A key provision of AILAD is the mandatory requirement for operators of high-risk AI systems to undergo regular third-party security and bias audits. These audits must be conducted by accredited, independent bodies to verify compliance with EU safety and fairness standards. The move is intended to increase transparency and accountability in the AI industry but has raised concerns among startups about the potential for high compliance costs. The directive will enter into force in late 2027, giving companies a transition period to adapt their development and deployment practices.