"Hoxhunt ChatGPT / Cybersecurity Research Reveals: Humans 1, AI 0"

Hoxhunt, the industry leader in behavior change software for cybersecurity, has issued a report analyzing the effectiveness of ChatGPT-generated phishing attempts. The study compared the success rate of simulated phishing attacks developed by human social engineers and those created by Artificial Intelligence (AI) Large Language Models (LLMs). While the potential for ChatGPT to be used for harmful phishing activities continues to captivate everyone's interest, Hoxhunt's research demonstrates that human social engineers continue to surpass AI when it comes to persuading users to click on dangerous links. The study found that professional red teamers generated a 4.2 percent click rate compared to ChatGPT's 2.9 percent click rate among an email user demographic sample. Sixty-nine percent of the time, humans outperformed AI at deceiving other humans. This article continues to discuss the effectiveness of ChatGPT-generated phishing attacks.

PR Newswire reports "Hoxhunt ChatGPT / Cybersecurity Research Reveals: Humans 1, AI 0"