Years Experience
CVEs Discovered
Public Speaking
Press Mentions
BUILDING
SECURE AI
Senior security researcher working at the intersection of LLM security, threat intelligence, and vulnerability research. I break AI systems the way real attackers do—from jailbreaks and prompt injection to LLM-powered malware and supply-chain backdoors—and turn those findings into practical defenses, threat reports, and education so organizations can safely adopt AI.
Core Expertise
HIGHLIGHTS
Immersive World Jailbreak
Discovered novel jailbreak technique affecting Microsoft Copilot, enabling bypass of AI safety guardrails.
HashJack Research
First known indirect prompt injection technique weaponizing URL fragments to manipulate AI browser assistants.
LAMEHUG Malware Analysis
First known LLM-powered malware with links to APT28 (Fancy Bear) threat actor.
Zero-Knowledge Threat Actor
Introducing the concept of attackers leveraging AI to bypass traditional skill requirements.
RECOGNITION & CVE DISCOVERIES
CVEs
CVE-2026-0859
Insecure Deserialization vulnerability in TYPO3 CMS Mailer file spool allowing local users with write access to the spool directory to inject and execute arbitrary PHP code via crafted serialized files.
ACHIEVEMENTS
AI Safety - Immersive world jailbreak
Microsoft
2025PRESS
The Rise of the Zero-Knowledge Threat Actor: New LLM Jailbreak Technique Discovered by Cato Networks Enables Easy Creation of Password-Stealing Malware
Press release quoting Vitaly Simonovich describing the Immersive World jailbreak that coerced GenAI tools into writing Chrome infostealers.
Security Researcher Proves GenAI Tools Can Develop Google Chrome Infostealers
Feature about your Cato CTRL experiment where you used Immersive World to get LLMs to write a Chrome infostealer.
AI Chatbots Can Be Easy Prey for 'Zero-Knowledge' Hackers
Article quoting you about Immersive World and describing how you bypassed safeguards in ChatGPT, Copilot and DeepSeek.
AI Chatbots: The New Target For Zero-Knowledge Hackers
Blog summarizing TechNewsWorld's coverage of your Immersive World jailbreak and naming you as the researcher.
When Fiction Becomes a Cyber Weapon: How AI Was Tricked into Writing Malware
Article citing Vitaly Simonovich by name and describing Immersive World as a new flavor of AI jailbreak.
Vibe Hacking: When AI Turns Beginners into Cybercriminals
Blog explicitly citing you and your Immersive World experiment as proof that non-experts can create malware with AI.
WormGPT returns: New malicious AI variants built on Grok and Mixtral uncovered
News article quoting Cato CTRL researcher Vitaly Simonovich on WormGPT variants that hijack mainstream LLM APIs.
Researchers say AI hacking tools sold online were powered by Grok, Mixtral
Story citing Cato Networks researcher Vitaly Simonovich about underground WormGPT services built on Grok and Mixtral.
The end of perimeter defense: When your own AI tools become the threat actor
Long-form piece built around your LAMEHUG research and Immersive World demo; includes extensive quotes from you.
AI Browsers Face A New Kind Of Attack, And It Puts Your Privacy At Risk
Consumer-focused coverage of HashJack attack demonstrating how AI browsers can display incorrect medication dosages on trusted pharmaceutical sites.
HashJack Attack Targets AI Browsers and Agentic AI Systems
Technical analysis of HashJack vulnerability with enterprise security recommendations for AI agents and agentic systems.
Zero-Click Agentic Browser Attack Can Delete Entire Google Drive Using Crafted Emails
The Hacker News coverage featuring HashJack attack by Cato Networks alongside other AI browser vulnerabilities, quoting Vitaly Simonovich.
HashJack Attack Uses URL '#' to Control AI Browser Behavior
Coverage of Cato CTRL's HashJack disclosure showing how URL fragments can be exploited to inject malicious prompts into AI browser assistants.
In Other News: HashJack AI Browser Attack, Charming Kitten Leak, Hacker Unmasked
SecurityWeek news roundup featuring HashJack AI browser attack discovered by Cato CTRL researchers.
AI browsers can be hacked with a simple hashtag, experts warn
TechRadar Pro coverage warning about HashJack technique that allows AI browsers to be hijacked with just a hashtag in a URL.
HashJack Indirect Prompt Injection Weaponizes Websites
Coverage of HashJack technique discovered by Cato CTRL that enables attackers to hide malicious instructions in URL fragments to manipulate AI browsers.
New 'HashJack' Attack Can Hijack AI Browsers and Assistants
Article on Cato Networks' HashJack research showing how URL fragments can be weaponized to force AI browsers to deliver phishing links or exfiltrate data.
HashJack: New Attack Technique Tricks AI Browsers Using a Simple '#'
Coverage of Cato CTRL's HashJack indirect prompt injection technique that exploits URL fragments to manipulate AI browser assistants.
HashJack: A Novel Exploit Leveraging URL Fragments To Deceive AI Browsers
Article on HashJack security threat that exploits how URL fragments function in AI-powered browsers to deceive and manipulate them.
Israeli tech firm finds AI vulnerability – Gemini susceptible
Coverage of Cato Networks' CTRL research group discovery of HashJack vulnerability affecting Google Gemini, Microsoft Copilot, and Perplexity Comet.
Password-Stealing AI HashJack Threat To Web Browsers Confirmed
Davey Winder covers Cato Networks' discovery of HashJack, a new AI browser threat that can steal passwords by exploiting URL fragments.
AI browser assistants vulnerable to HashJack prompt injection technique
Coverage of Cato CTRL's HashJack research showing AI browser assistants are vulnerable to indirect prompt injection via URL fragments.
HashJack: A New Attack That Fools AI Browsers With a Simple '#'
Coverage of HashJack vulnerability showing how URL fragments bypass traditional security defenses to manipulate AI browsers.
ИИ-браузеры можно обмануть с помощью символа «#»
Russian coverage of HashJack research quoting Vitaly Simonovich on how attackers can use URL fragments to inject hidden commands into AI browsers.
HashJack: новая атака через символ # обманывает ИИ-браузеры
Russian security news coverage of HashJack vulnerability affecting AI browser assistants like Gemini, Copilot, and Comet.
New 'HashJack' technique lets attackers manipulate AI assistants in Comet, Copilot and Gemini
Coverage of Cato CTRL's discovery of HashJack, an indirect prompt injection technique that exploits URL fragments to manipulate AI browser assistants.
HashJack attack shows AI browsers can be fooled with a simple '#'
Feature on Cato Networks' discovery of HashJack vulnerability that exploits URL fragments to manipulate AI browser assistants like Copilot, Gemini, and Comet.
How AI-powered hackers are stealing billions
Business feature that opens with your Velora/Jaxon story and names you as the Cato Networks researcher who tricked ChatGPT into writing malware.
Into the storm
Commentary that quotes Cato threat researcher Vitaly Simonovich and his Immersive World jailbreak technique.
Google Chrome Passwords Alert—Beware The Rise Of The AI Infostealers
Column quoting you on the Immersive World jailbreak and the risk of AI-generated infostealers targeting Chrome passwords.
How do you get ChatGPT to create malware strong enough to breach Google's password manager? Just play pretend.
Interview where you explain how narrative role-play (Immersive World) led ChatGPT to write a Chrome password-stealing infostealer.
Warning! Botnet is in your house...
Conference talk slide deck by Sarit Yerushalmi and Vitaly Simonovich (Imperva) on home-network botnet infections.
Mirai-Like Botnet Wages Massive Application-Layer DDoS Attack
Article quoting Imperva security researcher Vitaly Simonovich about a Mirai-like IoT botnet behind a massive L7 DDoS attack.
Massive Botnet Attack Used More Than 400000 IoT Devices
Coverage quoting you that the attack was the largest Layer-7 DDoS Imperva had ever seen.
Large-scale DDoS Attack Abuses HTML's Hyperlink Audit Ping Facility
Feature describing an attack monitored by Imperva researchers Vitaly Simonovich and Dima Bekerman using HTML5 ping for DDoS.
How HTML5 Ping Is Used in DDoS Attacks
Explainer quoting Imperva security researcher Vitaly Simonovich on how attackers abused HTML5 ping for application-layer DDoS.
Imperva Discovers New DDoS Attack that Abuses HTML5 Ping-Based Hyperlink Auditing Feature
News item crediting Imperva experts Vitaly Simonovich and Dima Bekerman for discovering the HTML5 ping DDoS technique.
Vulnerable Docker Hosts Actively Abused in Cryptojacking Campaigns
News article describing how Imperva's Vitaly Simonovich and Ori Nakar found thousands of exposed Docker daemons abused for Monero mining.
Imperva Researchers Find Hundreds of Vulnerable Docker Hosts Exploited by Cryptominers
Coverage of Imperva research where Vitaly Simonovich and Ori Nakar show attackers abusing exposed Docker APIs to run cryptominers.
Docker API vulnerability allows hackers to mine Monero
Article quoting Imperva researchers Vitaly Simonovich and Ori Nakar on misconfigured Docker APIs being hijacked for Monero mining.
The Rise of the Zero-Knowledge Threat Actor: New LLM Jailbreak Technique Discovered by Cato Networks Enables Easy Creation of Password-Stealing Malware
Press release quoting Vitaly Simonovich describing the Immersive World jailbreak that coerced GenAI tools into writing Chrome infostealers.
Security Researcher Proves GenAI Tools Can Develop Google Chrome Infostealers
Feature about your Cato CTRL experiment where you used Immersive World to get LLMs to write a Chrome infostealer.
AI Chatbots Can Be Easy Prey for 'Zero-Knowledge' Hackers
Article quoting you about Immersive World and describing how you bypassed safeguards in ChatGPT, Copilot and DeepSeek.
AI Chatbots: The New Target For Zero-Knowledge Hackers
Blog summarizing TechNewsWorld's coverage of your Immersive World jailbreak and naming you as the researcher.
When Fiction Becomes a Cyber Weapon: How AI Was Tricked into Writing Malware
Article citing Vitaly Simonovich by name and describing Immersive World as a new flavor of AI jailbreak.
Vibe Hacking: When AI Turns Beginners into Cybercriminals
Blog explicitly citing you and your Immersive World experiment as proof that non-experts can create malware with AI.
WormGPT returns: New malicious AI variants built on Grok and Mixtral uncovered
News article quoting Cato CTRL researcher Vitaly Simonovich on WormGPT variants that hijack mainstream LLM APIs.
Researchers say AI hacking tools sold online were powered by Grok, Mixtral
Story citing Cato Networks researcher Vitaly Simonovich about underground WormGPT services built on Grok and Mixtral.
The end of perimeter defense: When your own AI tools become the threat actor
Long-form piece built around your LAMEHUG research and Immersive World demo; includes extensive quotes from you.
AI Browsers Face A New Kind Of Attack, And It Puts Your Privacy At Risk
Consumer-focused coverage of HashJack attack demonstrating how AI browsers can display incorrect medication dosages on trusted pharmaceutical sites.
HashJack Attack Targets AI Browsers and Agentic AI Systems
Technical analysis of HashJack vulnerability with enterprise security recommendations for AI agents and agentic systems.
Zero-Click Agentic Browser Attack Can Delete Entire Google Drive Using Crafted Emails
The Hacker News coverage featuring HashJack attack by Cato Networks alongside other AI browser vulnerabilities, quoting Vitaly Simonovich.
HashJack Attack Uses URL '#' to Control AI Browser Behavior
Coverage of Cato CTRL's HashJack disclosure showing how URL fragments can be exploited to inject malicious prompts into AI browser assistants.
In Other News: HashJack AI Browser Attack, Charming Kitten Leak, Hacker Unmasked
SecurityWeek news roundup featuring HashJack AI browser attack discovered by Cato CTRL researchers.
AI browsers can be hacked with a simple hashtag, experts warn
TechRadar Pro coverage warning about HashJack technique that allows AI browsers to be hijacked with just a hashtag in a URL.
HashJack Indirect Prompt Injection Weaponizes Websites
Coverage of HashJack technique discovered by Cato CTRL that enables attackers to hide malicious instructions in URL fragments to manipulate AI browsers.
New 'HashJack' Attack Can Hijack AI Browsers and Assistants
Article on Cato Networks' HashJack research showing how URL fragments can be weaponized to force AI browsers to deliver phishing links or exfiltrate data.
HashJack: New Attack Technique Tricks AI Browsers Using a Simple '#'
Coverage of Cato CTRL's HashJack indirect prompt injection technique that exploits URL fragments to manipulate AI browser assistants.
HashJack: A Novel Exploit Leveraging URL Fragments To Deceive AI Browsers
Article on HashJack security threat that exploits how URL fragments function in AI-powered browsers to deceive and manipulate them.
Israeli tech firm finds AI vulnerability – Gemini susceptible
Coverage of Cato Networks' CTRL research group discovery of HashJack vulnerability affecting Google Gemini, Microsoft Copilot, and Perplexity Comet.
Password-Stealing AI HashJack Threat To Web Browsers Confirmed
Davey Winder covers Cato Networks' discovery of HashJack, a new AI browser threat that can steal passwords by exploiting URL fragments.
AI browser assistants vulnerable to HashJack prompt injection technique
Coverage of Cato CTRL's HashJack research showing AI browser assistants are vulnerable to indirect prompt injection via URL fragments.
HashJack: A New Attack That Fools AI Browsers With a Simple '#'
Coverage of HashJack vulnerability showing how URL fragments bypass traditional security defenses to manipulate AI browsers.
ИИ-браузеры можно обмануть с помощью символа «#»
Russian coverage of HashJack research quoting Vitaly Simonovich on how attackers can use URL fragments to inject hidden commands into AI browsers.
HashJack: новая атака через символ # обманывает ИИ-браузеры
Russian security news coverage of HashJack vulnerability affecting AI browser assistants like Gemini, Copilot, and Comet.
New 'HashJack' technique lets attackers manipulate AI assistants in Comet, Copilot and Gemini
Coverage of Cato CTRL's discovery of HashJack, an indirect prompt injection technique that exploits URL fragments to manipulate AI browser assistants.
HashJack attack shows AI browsers can be fooled with a simple '#'
Feature on Cato Networks' discovery of HashJack vulnerability that exploits URL fragments to manipulate AI browser assistants like Copilot, Gemini, and Comet.
How AI-powered hackers are stealing billions
Business feature that opens with your Velora/Jaxon story and names you as the Cato Networks researcher who tricked ChatGPT into writing malware.
Into the storm
Commentary that quotes Cato threat researcher Vitaly Simonovich and his Immersive World jailbreak technique.
Google Chrome Passwords Alert—Beware The Rise Of The AI Infostealers
Column quoting you on the Immersive World jailbreak and the risk of AI-generated infostealers targeting Chrome passwords.
How do you get ChatGPT to create malware strong enough to breach Google's password manager? Just play pretend.
Interview where you explain how narrative role-play (Immersive World) led ChatGPT to write a Chrome password-stealing infostealer.
Warning! Botnet is in your house...
Conference talk slide deck by Sarit Yerushalmi and Vitaly Simonovich (Imperva) on home-network botnet infections.
Mirai-Like Botnet Wages Massive Application-Layer DDoS Attack
Article quoting Imperva security researcher Vitaly Simonovich about a Mirai-like IoT botnet behind a massive L7 DDoS attack.
Massive Botnet Attack Used More Than 400000 IoT Devices
Coverage quoting you that the attack was the largest Layer-7 DDoS Imperva had ever seen.
Large-scale DDoS Attack Abuses HTML's Hyperlink Audit Ping Facility
Feature describing an attack monitored by Imperva researchers Vitaly Simonovich and Dima Bekerman using HTML5 ping for DDoS.
How HTML5 Ping Is Used in DDoS Attacks
Explainer quoting Imperva security researcher Vitaly Simonovich on how attackers abused HTML5 ping for application-layer DDoS.
Imperva Discovers New DDoS Attack that Abuses HTML5 Ping-Based Hyperlink Auditing Feature
News item crediting Imperva experts Vitaly Simonovich and Dima Bekerman for discovering the HTML5 ping DDoS technique.
Vulnerable Docker Hosts Actively Abused in Cryptojacking Campaigns
News article describing how Imperva's Vitaly Simonovich and Ori Nakar found thousands of exposed Docker daemons abused for Monero mining.
Imperva Researchers Find Hundreds of Vulnerable Docker Hosts Exploited by Cryptominers
Coverage of Imperva research where Vitaly Simonovich and Ori Nakar show attackers abusing exposed Docker APIs to run cryptominers.
Docker API vulnerability allows hackers to mine Monero
Article quoting Imperva researchers Vitaly Simonovich and Ori Nakar on misconfigured Docker APIs being hijacked for Monero mining.
RESEARCH18
Cato CTRL Threat Research: HashJack - First Known Indirect Prompt Injection Against AI Browser Assistants
Novel indirect prompt injection technique that weaponizes URL fragments to manipulate AI browser assistants like Copilot, Gemini, and Comet.
SPEAKING8
The Anatomy of Criminal Failure: Analyzing OpSec Flaws in Major Takedowns
An in-depth analysis of operational security failures that led to the takedown of major cybercriminal operations, examining the mistakes that ultimately exposed threat actors to law enforcement.
MY BLOG10
Threat Actors Don't Care About Your Fucking Scope
A critical examination of how bug bounty scope limitations create dangerous security blind spots, with insights from Microsoft's shift to 'In Scope By Default' and real-world breach examples.
Get in touch
Lectures, interviews, or collaborations?
Always open to discussing new opportunities, interesting projects, or just chatting about security and AI.