AI Tilts Cyber Race Toward Attackers, Defenders Lag
Former FBI cyber leader Cynthia Kaiser says defenders are still catching up to attackers using artificial intelligence.
Updated on Nov. 2, 2026 in Cybersecurity
Live Poll
Do you trust that a familiar voice on the phone is really who it claims to be?
Former FBI cyber leader Cynthia Kaiser, who now leads ransomware research at Halcyon, assessed on November 2, 2026, that attackers are currently using AI slightly more than defenders. She warned that AI makes phishing and social engineering, including voice phishing, far more convincing.
Why it matters
AI lowers the effort needed to produce convincing messages and manipulate people, which raises risk for organizations of every size. Small and medium-sized enterprises can serve as entry points into larger organizations through supply chains, extending the reach of these threats.
The assessment rests on expert judgment rather than a published dataset, with the most concrete figure being that as little as three seconds of audio can be used to clone a voice. The source does not quantify the size of the AI usage gap or confirm how quickly defenders will close it.
The players
Cynthia Kaiser
Former FBI cyber leader and current head of ransomware research at Halcyon, a cybersecurity firm, known for her work on attacker tactics and defense strategy.
Halcyon
A cybersecurity company where Cynthia Kaiser leads ransomware research, focused on helping organizations detect and respond to ransomware threats.
The details
Cynthia Kaiser made the remarks during a special joint episode of the podcasts Ctrl Alt Lead and Ready Set Mid-Market. She backs layered verification, including multi-factor authentication, rather than trusting perceived voice identity, and she noted that nation-state actors may dedicate small teams to experimenting with AI models for offensive use.
Timeline
November 2, 2026: Kaiser's assessment on AI and cyber defense is recorded in the joint podcast episode.
Big Picture
Kaiser's assessment draws on her years leading cyber investigations at the FBI, which shaped how ransomware threats were understood at the national level. Her current view that attackers hold a slight edge in AI adoption extends that record into a period where generative tools are reshaping phishing and voice fraud.
Individuals and small businesses should not rely on recognizing a familiar voice as proof of identity, since AI can clone a voice from very little audio. Adding multi-factor authentication and verifying requests through a second channel are practical steps that reduce exposure to convincing phishing and voice fraud.
The takeaway
Treat any urgent request that arrives by voice or message as unverified until confirmed through a separate channel. Layering checks such as multi-factor authentication protects against the increasingly convincing social engineering that AI enables.
Further reading
For more on how these threats evolve, explore the Cybersecurity section.
Source note: This article includes information reported by Computing.
Live Poll
Do you trust that a familiar voice on the phone is really who it claims to be?
