AI Agents Are Infiltrating Group Chats: The Internet Can No Longer Tell Who Is Human

The rise of the "agentic web," where autonomous AI agents perform tasks, interact, and even impersonate humans, has rendered the internet’s foundational assumption of human-only participation obsolete. Academic and policy researcher Renee DiResta highlights how these agents can now engage in sophisticated social engineering, such as gaslighting users or spreading malware, necessitating a shift toward robust "proof of personhood" systems. While traditional verification methods like government-issued digital IDs, biometrics, and hardware tokens offer potential solutions, they introduce significant privacy risks and centralization concerns. Instead, a more viable path involves context-aware credentials and zero-knowledge proofs that verify humanity without compromising user anonymity. As agents become increasingly ubiquitous, establishing clear accountability for their actions—while balancing the need for privacy—remains a critical challenge for maintaining the integrity of online spaces.