AI Agents to Halve Account Exploitation Time by 2027 Predicts Gartner

By 2027, AI agents will reduce the time it takes to exploit account exposures by 50%, according to Gartner, Inc. This alarming prediction highlights the growing threat of automated cyberattacks leveraging weak authentication methods.

The Rise of Automated Account Takeovers

Account takeover (ATO) remains a persistent attack vector due to weak authentication credentials like passwords, which are often gathered through data breaches, phishing, social engineering, and malware. Jeremy D'Hoinne, VP Analyst at Gartner, explains, "Attackers then leverage bots to automate login attempts across various services, hoping credentials have been reused."

AI agents will further automate ATO steps, from deepfake-based social engineering to end-to-end credential abuse. This automation will make attacks faster and more scalable.

Industry Response and Defense Strategies

In response, vendors are expected to introduce products for web, app, API, and voice channels to detect and monitor AI agent interactions. Akif Khan, VP Analyst at Gartner, advises, "Security leaders should expedite the move toward passwordless, phishing-resistant MFA. For customer use cases, educate and incentivize users to migrate from passwords to multidevice passkeys."

The Growing Threat of Social Engineering

Gartner also predicts that by 2028, 40% of social engineering attacks will target executives and the broader workforce. Attackers are combining traditional tactics with counterfeit reality techniques like deepfake audio and video to deceive employees during calls.

While few high-profile cases have been reported, these incidents have resulted in significant financial losses. Detecting deepfakes remains challenging, especially in real-time communications across diverse platforms.

Organizational Preparedness

Manuel Acosta, Sr. Director Analyst at Gartner, emphasizes, "Organizations must adapt procedures and workflows to resist attacks leveraging counterfeit reality techniques. Educating employees about social engineering with deepfakes is critical."

Key Takeaways:

• AI agents will automate 50% of account exploitation by 2027.