Charm Security
Agentic AI workforce that investigates and intervenes on scams and fraud in real time, reading manipulation and intent behind fraud signals.
Visit Website ↗ + Add to CompareOverview
Charm Security builds an “agentic AI workforce” for scam and fraud defense that goes beyond detection to active intervention, combining fraud expertise, behavioral science, and AI agents that assess a customer’s Human Vulnerability and Exposure (HVE) profile in real time. Rather than only flagging suspicious transactions, Charm’s agents attempt to intervene with the customer directly during a scam in progress, before the loss completes.
Founded by Roy Zur, a third-time cybersecurity entrepreneur who previously built Cybint (now ThriveDX), Charm is backed by Israeli venture studio Team8 and was named a 2026 RSAC Innovation Sandbox company. In March 2026 it partnered with Reality Defender to embed real-time deepfake detection for voice, image, and text into its agentic workforce, extending its scam-defense coverage into AI-generated fraud.
Innovation Matrix Assessment
Moved quickly from Team8 incubation to an RSAC Innovation Sandbox selection and a Reality Defender deepfake-detection integration within a short window.
Real-time intervention during an in-progress scam is a meaningfully more useful capability for fraud teams than post-hoc transaction flagging alone.
RSAC Innovation Sandbox selection and a named integration partner (Reality Defender) are credible early momentum signals for a young company. Recognized in Cyber Defense Media Group's 2026 Global InfoSec Awards (2 awards), independently juried industry validation of market traction.
Framing fraud defense as an agentic AI workforce that actively engages the victim in real time is a distinct approach from traditional rules-based fraud scoring.
The behavioral-science and HVE-profiling approach is conceptually strong but lacks published independent results on actual fraud-loss reduction.
AI-generated scams and deepfake-enabled fraud are a rapidly growing threat category that will only intensify over the next several years.
Why CISOs Should Care
Addresses the fastest-growing fraud vector -- AI-generated social engineering -- with active, real-time intervention rather than after-the-fact alerts.
What Makes It Different
Agentic AI that actively engages the targeted customer mid-scam, combined with deepfake detection via its Reality Defender partnership.
The Matrix Verdict
75/100 — MEANINGFUL INNOVATOR
An ambitious, well-pedigreed early-stage entrant tackling AI-era fraud with a genuinely novel intervention model; still needs a track record.
Editorial Note: Claims vs. Verified Findings
Efficacy and intervention-success claims are vendor and partner-published; no independent fraud-reduction study was located.
Sources
Alternatives to Charm Security
Adaptive Security
AI-driven platform that simulates deepfake, voice, and multichannel social-engineering attacks to train and test organizations against next-generation phishing.
Quilr
Early-stage agentic AI security startup building a 'Service-as-Software' platform to guard against human-related breaches and secure AI agent…
Zenity
Governance and security platform for AI agents and low-code/no-code development, securing agent identity, permissions and behavior across the…
Tenzai
An agentic AI penetration testing startup building autonomous 'AI hackers' to find and validate exploitable vulnerabilities at a…
Reco
Reco secures the "agentic ecosystem" — mapping what AI agents can access across SaaS and enterprise apps, detecting…
Alice (formerly ActiveFence)
Israeli AI security company (rebranded from ActiveFence in January 2026) offering a lifecycle platform to test, guard, and…