Improving online anti-phishing training using cognitive large language models

Open Webpage

Tailia Malloy, Fei Fang, Cleotilde Gonzalez

Published: 01 Nov 2025, Last Modified: 25 Nov 2025Computers in Human BehaviorEveryoneRevisionsCC BY-SA 4.0
Abstract: Highlights•Examples written by humans and stylized with code from GPT-4 were the most difficult.•Our Cognitive Large Language Model showed the best learning outcomes for students.•This CLLM selected educational examples and informed natural language feedback.•We suggest LLMs for education should be informed of human cognitive mechanisms.