The Register: Nearly half of businesses suffered deepfaked phone calls against staff

Sep 23, 2025

—

Source URL: https://www.theregister.com/2025/09/23/gartner_ai_attack/
Source: The Register
Title: Nearly half of businesses suffered deepfaked phone calls against staff

Feedly Summary: AI attacks on the rise
A survey of cybersecurity bosses has shown that 62 percent reported attacks on their staff using AI over the last year, either by the use of prompt injection attacks or faking out their systems using phony audio or video generated by AI.…

AI Summary and Description: Yes

Summary: The reported rise in AI attacks, particularly through techniques like prompt injection and the use of faked audio or video, underscores the urgent need for enhanced cybersecurity measures. This survey reveals critical insights for security professionals in understanding the vulnerabilities posed by AI technologies.

Detailed Description: The text discusses a significant increase in AI-driven attacks on organizations, highlighting the evolving landscape of cybersecurity threats. The findings from a survey of cybersecurity executives indicate that professionals must adapt to these new types of attacks, which leverage the capabilities of AI technologies.

– **Survey Results**: 62% of cybersecurity leaders noted experiencing AI-related attacks in the past year.
– **Types of Attacks**:
– **Prompt Injection Attacks**: These attacks manipulate AI systems by inserting deceptive inputs, leading to potentially harmful outputs.
– **Faked Audio and Video**: Attackers are using generative AI to create realistic but false audio and video, posing serious risks to organizational security.

The increased prevalence of these attacks suggests a need for:

– **Security Framework Development**: Organizations should look to enhance their security measures by incorporating protocols that specifically address AI vulnerabilities.
– **Training and Awareness**: Employees must be educated about these new threats and trained to recognize deceptive content.
– **Advanced Detection Mechanisms**: Implementing systems capable of detecting AI-generated false information will be essential in maintaining security integrity.

Overall, the document serves as a crucial reminder for cybersecurity professionals to stay proactive in safeguarding technologies against the rapidly evolving tactics of cyber attackers, particularly those leveraging artificial intelligence.

2 2025 3 5 a Act advanced advanced detection advanced detection mechanisms age AGI AI AI systems AI technologies All and API ARM art artificial Artificial Intelligence as at ated attack attacker attackers attacks audio aware awareness Bi business by C capabilities CI CIA co Col content core critical cyber cyber attack cyber attackers cybersecurit Cybersecurity cybersecurity measures Cybersecurity Professionals cybersecurity threat cybersecurity threats D de deceptive content deep deepfake detection detection mechanisms development document drive driven Driven Attacks e executives exp faked audio and video faking false information for framework g Gartner Gen generated generative Generative AI GIS gs H harm high Highlight HR http HTTPS in information injection injection attacks insights integrity Intel intelligence io J k l land leading led Li M man measures N new no NPU o of on one ons OPM organization organizational security organizations oS oss out output Outputs over per potential pre pro proactive professionals prompt prompt injection attack prompt injection attacks protocol protocols ps Q R rag rate RCE re real red report Risk risks RMF Ro RoT s safe sec security security framework security framework development security leaders security measure security measures security professionals security threat security threats Sig source specific SSE system systems T tactics tech techniques technologies ted text the threat threats to TP trained training training and awareness type under US use V val video vulnerabilities Ware Wi x z