ByJERUSALEM POST STAFF

JUNE 8, 2026 16:08

Updated: JUNE 8, 2026 16:09

Israeli-founded cybersecurity startup A Security left stealth mode on Monday after raising $37 million to grow its autonomous offensive security and remediation platform, according to the firm.

The round was led by Lightspeed Venture Partners along with Cyberstarts, Cerca Partners, and angel backers such as Wiz CEO Assaf Rappaport and Cyera CEO Yotam Segev.

A Security plans to deploy the new capital toward scaling operations and broadening a platform that identifies and patches exploit routes before malicious actors can leverage them.

The company enters the arena as security teams confront fresh risks from artificial intelligence-enabled attackers capable of locating flaws, linking them, and automating segments of cyber campaigns at speeds exceeding conventional human-led efforts.

"AI has altered the pace and reach of offensive cyber operations," noted Yossi Torati, CEO and co-founder of A Security. "Attackers now apply agentic tools to locate, chain, and exploit weaknesses across environments faster than human teams using manual methods can respond."

Torati explained that the firm was created to uncover "real-world exploit paths" that AI-enabled attackers would pursue to trigger breaches, outages, and operational harm, then assist organizations in eliminating those paths in advance.

Israeli cyber veterans behind the company

Prior to establishing A Security, Torati accumulated nearly two decades of experience advising global enterprises on cyberattacks and resilience, most recently serving as director of enterprise security at incident response firm Sygnia.

Co-founders Omer Gull and Yuval Itzchakov contribute backgrounds from Check Point Software, Hunters, and the IDF Unit 8200, the company stated.

A Security positions its platform as distinct from periodic penetration tests and manual risk reviews by delivering continuous detection of cross-domain attack paths and confirming whether they are truly exploitable.

The system deploys offensive and defensive AI agents to evaluate environments under controlled execution with complete audit trails, then supports remediation at the root cause and via additional security controls, the firm added.

Investors see growing AI cyber threat

"Today cyber defenders require a tool that can outpace weaponized AI," stated Guru Chahal, partner at Lightspeed. "We view A Security as that solution."

Hila Zigman, general partner at Cyberstarts, observed that the security sector underestimates the extent to which AI is reshaping offensive cyber operations.

"Attackers already function with autonomous capabilities at speeds and scales for which defenders were never prepared," Zigman said.

The company reported it is already engaged with major organizations in finance, healthcare, critical infrastructure, and technology, where teams must validate and close attack paths more rapidly than traditional testing allows.

The launch occurs amid rising industry concern over AI systems that aid vulnerability discovery and exploitation. Reuters noted last week that Anthropic is widening access to Claude Mythos Preview, an AI tool built to spot software flaws, for roughly 200 organizations via its Project Glasswing effort.

A Security said its platform targets organizations "where failure is not an option," including sectors unable to tolerate operational disruption from cyberattacks.