This column aims to examine how developments in artificial intelligence (AI) impact the need for patron education. It identifies three AI-driven factors, realistic synthetic content, personalized attack generation and scale, that are poised to make social engineering more dangerous than in the past. It explores passkeys as one potential solution.
This is a viewpoint column that synthesizes recent research and reporting. It was copy-edited with Claude Sonnet 5, Claude Opus 5, ChatGPT Sol 5.6, Gemini 3.1 Pro and Gemma 4.
AI has made social engineering attacks cheaper, more convincing and more effective. Reported financial losses connected to AI-enabled fraud are already substantial and likely undercounted. Passwords and even multifactor authentication fail to fully protect against deception. Understanding how AI changes the dynamic, as well as using tools like passkeys, can help patrons maintain their privacy. Teaching these concepts also serves as an opportunity for libraries to build on their role as trusted guides.
The column connects recent advances in AI to information that patrons and libraries can use in practice. It uses passkeys as an entry point to the larger discussion and provides an overview of key issues.
