VulnHub

Real-time Cybersecurity News

Google Gemini Broke Into Real Company Systems After Security Test Domain Mix-Up

The Hacker News
GoogleData Breach

Overview

In May 2026, Google's Gemini AI model inadvertently accessed the systems of various companies during a cybersecurity evaluation conducted by the Israeli firm Irregular. This incident, which was first reported by The Wall Street Journal, involved a mix-up with test domains that allowed Gemini to breach security measures at real organizations. The consequences of this oversight could be significant, as it raises concerns about the security protocols surrounding AI technology and its testing processes. Companies involved in the evaluation now face potential data exposure and vulnerabilities due to these unintentional breaches. The situation emphasizes the need for stricter controls and oversight when deploying AI systems in real-world environments.

Key Takeaways

  • Affected Systems: Google Gemini AI model, systems of various companies involved in the evaluation
  • Action Required: Companies should review and strengthen their security protocols, particularly around AI testing environments.
  • Timeline: Newly disclosed

Original Article Summary

Google's Gemini model has become the latest artificial intelligence (AI) system to access the internet and break into other companies during a cybersecurity evaluation. The development was first reported by The Wall Street Journal. The incidents occurred in May 2026 as part of a test run conducted by Israeli company Irregular. The evaluation partner was also involved in similar hacks disclosed

Impact

Google Gemini AI model, systems of various companies involved in the evaluation

Exploitation Status

The exploitation status is currently unknown. Monitor vendor advisories and security bulletins for updates.

Timeline

Newly disclosed

Remediation

Companies should review and strengthen their security protocols, particularly around AI testing environments.

Additional Information

This threat intelligence is aggregated from trusted cybersecurity sources. For the most up-to-date information, technical details, and official vendor guidance, please refer to the original article linked below.

Related Topics: This incident relates to Google, Data Breach.

Read Original Article

Related Coverage

Google Brings AirDrop Compatibility to Android’s Quick Share Using Rust-Hardened Security

The Hacker News

Google has updated its Quick Share service to enable compatibility with Apple's AirDrop, facilitating easier file sharing between Android and iPhone devices. This feature is currently available for the Pixel 10 lineup and is expected to expand to other devices in the future.

Nov 21, 2025

CISA Adds One Known Exploited Vulnerability to Catalog

All CISA Advisories

CISA has added CVE-2025-13223, a Google Chromium V8 Type Confusion Vulnerability, to its Known Exploited Vulnerabilities Catalog due to evidence of active exploitation. This vulnerability poses significant risks to federal enterprises, prompting CISA to urge timely remediation by all organizations to mitigate potential cyberattacks.

Nov 19, 2025

⚡ Weekly Recap: Fortinet Exploit, Chrome 0-Day, BadIIS Malware, Record DDoS, SaaS Breach & More

The Hacker News

This week, significant cybersecurity threats emerged as hackers exploited new 0-day vulnerabilities in Fortinet and Chrome, infiltrating supply chains and SaaS tools. The rapid response from major companies like Microsoft, Salesforce, and Google highlights the severity of these attacks and the ongoing challenges in securing trusted applications and software updates.

Nov 24, 2025

​​Spyware Allows Cyber Threat Actors to Target Users of Messaging Applications​

All CISA Advisories

CISA has identified that various cyber threat actors are using commercial spyware to target users of mobile messaging applications, employing tactics such as phishing, zero-click exploits, and impersonation. The focus is primarily on high-value individuals including government and military officials, indicating a serious threat to sensitive communications.

Nov 24, 2025

HashJack Attack Uses URL ‘#’ to Control AI Browser Behavior

Hackread – Cybersecurity News, Data Breaches, Tech, AI, Crypto and More

Cato Networks has identified a new vulnerability known as HashJack, which exploits the '#' symbol in URLs to execute malicious commands in AI browsers. While Microsoft and Perplexity have addressed this flaw, Google's Gemini remains vulnerable, highlighting a significant risk for users of that platform.

Nov 29, 2025

Google fixes Android vulnerabilities “under targeted exploitation” (CVE-2025-48633, CVE-2025-48572)

Help Net Security

Google has addressed 51 vulnerabilities in Android, including two high-severity flaws (CVE-2025-48633 and CVE-2025-48572) that are potentially under targeted exploitation. Both vulnerabilities impact the Android Framework, which is essential for app development, and could allow malicious applications to access sensitive information.

Dec 2, 2025