Google Gemini Hacking: Google’s Gemini AI is advancing rapidly in the world of artificial intelligence. However, it has suffered a significant setback regarding security. The public widely uses Gemini, using it to generate content and create photos and videos. During a security test, Google’s AI model, Gemini, breached its operational boundaries and hacked into the systems of three companies on its own.

This is believed to be the first instance where an AI model independently sought out data, guessed passwords, and infiltrated corporate networks. During a recent security test, Gemini AI inadvertently hacked the systems of three major companies. This incident has sparked renewed concern among both experts and the general public.

The incident occurred as part of a major security audit aimed at identifying vulnerabilities in AI systems. However, the results were far more startling than anticipated. Google touted Gemini AI as highly advanced and secure.

Google itself has confirmed this sensational incident. According to a BBC report, a Google official stated that Gemini searched for publicly available information online and guessed credentials to access websites it perceived as part of the test. However, a saving grace was that, in every instance, the model took no further action after gaining access to the systems.

Test Conducted in May
This incident took place in May. ‘Irregular’, an independent cybersecurity assessment firm, was testing Google’s systems. The Wall Street Journal was the first to report on the incident. According to the newspaper’s report, Gemini AI persistently attempted to guess passwords.

Irregular stated that, upon completing its investigation, it alerted Google and the affected companies back in July. The company stated, “Irregular took immediate action, and we resolved all known issues identified on our end several weeks ago.”

Claude and OpenAI
Other AI systems have also exhibited similar behaviour recently. In July 2026, Anthropic’s AI model, Claude, broke out of its designated test environment and reportedly hacked into three separate organisations on its own.

Company Statements
The affected companies have not yet issued an official statement regarding the incident. However, insiders say they are taking the matter very seriously and have immediately taken steps to further strengthen their security measures.