In a stark warning for enterprise security, especially for firms that handle sensitive personal and financial information, a recent study indicates that 88% of organizations experienced at least one security incident that undermined trust in digital communications over the past year. The rise of AI-powered phishing attacks has catalyzed a troubling resurgence of threats that legacy security tools are ill-equipped to counter.
The research report, conducted by Osterman Research and commissioned by IRONSCALES, surveyed 128 cybersecurity decision-makers and uncovered a perilous gap in preparedness: while 82% of respondents noted an increase in threat actors looking to exploit trusted communications, 60% lack confidence in their ability to effectively counter deepfake attacks.
Michael Sampson, Principal Analyst at Osterman Research, remarked, “The threat curve just got reset. Even ‘solved’ attack types like phishing and business email compromise have become immature again. BEC attacks from 2025 bear little resemblance to those from 2020—they’re now hyper-personalized, multi-channel, and can be launched autonomously at scale.” This rising complexity in attacks further complicates an already dire landscape, as organizations grapple with high breach rates.
Despite these challenges, respondents believe that the maturity of AI-enhanced attacks is still developing. Specifically, 28% indicate that AI-generated phishing is just beginning to emerge, alongside 25% who feel the same about deepfake audio attacks, and another 28% who consider deepfake video attacks to be in nascent stages. In essence, organizations are facing alarming breach rates with threats that have not yet reached their full potential.
Traditional indicators that employees and security systems previously relied on—such as grammar errors, suspicious sender addresses, and generic language—have been rendered ineffective by AI. Cybercriminals can now craft meticulous attacks in any language, with personalization occurring at scale. These attacks are increasingly delivered via multiple channels, including email, phone, video, and collaboration platforms.
The research highlights a “perfect storm” of vulnerability for finance departments, which are viewed as the highest-priority target for threat actors. A significant 59% of organizations classify finance teams as “high” or “extreme” priority targets, while the same percentage expresses substantial concern about these teams’ readiness to defend against trust-based attacks. Audian Paxson, Principal Technical Strategist at IRONSCALES, emphasized, “Finance teams control the money, so they’re priority number one for attackers. But cybersecurity leaders report the lowest confidence in these teams’ ability to spot sophisticated BEC and impersonation scams. That gap is getting exploited daily.”
Moreover, over 33% of organizations reported that threat actors successfully impersonated trusted vendors to steal funds or information over the past year, with vendor impersonation attacks showing a significant increase—13% of respondents noted major year-over-year growth in such incidents.
Perhaps most alarmingly, nearly one in five security leaders stated that security awareness training has proven ineffective against AI-enhanced threats. Current training methods aimed at preparing employees to detect trust-exploiting attacks are failing many organizations, particularly when it comes to detecting deepfake audio and video attacks. Respondents rated the effectiveness of their training as follows: 38% for detecting deepfake audio attacks, 39% for deepfake video attacks, and 43% for AI-generated phishing.
“The legacy email protections are too blunt an instrument to recognize the subtle indicators of modern AI-powered attacks,” noted Sampson. “Organizations can no longer trust these legacy solutions to protect against threats that didn’t exist when they were designed.”
The growing crisis is prompting a reassessment of security strategies across organizations. The research found that 70% of organizations now consider detecting deepfake audio impersonation attacks “extremely important,” marking the highest priority increase among respondents. Additionally, 70% are willing to integrate best-in-class point solutions to address existing gaps, 68% are open to changing vendors entirely, and 70% are prepared to replace their entire security technology stack.
The cost of inaction is becoming increasingly clear. Fifty-five percent of security leaders believe that failing to defend against trust-exploiting attacks significantly heightens the likelihood of data breaches, with the fallout extending to reduced productivity, compromised customer communications, and broader operational disruption. As organizations navigate this evolving threat landscape, the imperative for robust, future-ready security measures has never been more urgent.
See also
Anthropic’s Claims of AI-Driven Cyberattacks Raise Industry Skepticism
Anthropic Reports AI-Driven Cyberattack Linked to Chinese Espionage
Quantum Computing Threatens Current Cryptography, Experts Seek Solutions
Anthropic’s Claude AI exploited in significant cyber-espionage operation
AI Poisoning Attacks Surge 40%: Businesses Face Growing Cybersecurity Risks
