EU AI Act Imposes Risk-Based Rules for Trustworthy AI Launching August 2024

AI agents are rapidly transforming industries by autonomously performing complex tasks, from e-commerce to healthcare and cybersecurity, but their development presents significant challenges and risks. Forter introduced new trust platform features to detect and manage AI agent activity in e-commerce, responding to a surge in AI-driven interactions that complicate fraud prevention and customer experience. Meanwhile, leading AI researchers at the Agentic AI Summit highlighted current limitations in AI agent reliability, safety, and trustworthiness, emphasizing the gap between promising demonstrations and practical deployment. The European Union’s Artificial Intelligence Act, effective August 2024, establishes a risk-based regulatory framework to ensure accountability and transparency for AI systems, particularly those deemed high-risk, influencing global AI governance standards. Developers like Anthropic are proposing frameworks focused on balancing agent autonomy with human oversight to ensure AI agents operate safely, reliably, and in alignment with human values. Across sectors, implementing safety guardrails—such as clear operational boundaries, compliance with regulations like HIPAA, and transparent communication of agent limitations—is critical to maintaining trust and mitigating ethical and security risks associated with autonomous agents.