Google’s Gemini Reportedly Accesses Internet and Hacks 3 Other Companies in the First Known Breakout of the Company’s AI Mode


Google’s artificial intelligence model, better known as Gemini, reportedly hacked three different companies during a cybersecurity test run.

The AI model is already wreaking havoc (Image via Google)
The AI model is already wreaking havoc (Image via Google)

Google’s artificial intelligence model, better known as Gemini, reportedly hacked three companies. This is the first recorded breakout, and it occurred during a test run to see its cybersecurity capabilities. The test model accidentally gained access to the internet and pursued real companies instead of fake ones. After assessing the situation, Google’s AI tool shut itself down.

Around May 2026, Google confirmed its AI model gained unauthorized access to three companies. The AI model broke through using leaked online credentials or through brute force and guessing. Heather Adkins, the vice president of security engineering at Google, shared more details about the reported hacks.

“In a standard evaluation, the model found public information online and guessed credentials to access websites it thought were part of the test,”

She later added that Gemini stopped itself before it could do more damage, and this incident encouraged the company to be more vigilant about what the AI tool can do.

“In all three of these instances, the model stopped.”

Read More: “Some People Like Where They Live”: Popular Streamer Explains Why He Won’t Buy His Parents a New Home Despite Being a Millionaire

Google Gemini Hacked Real Companies Instead of Fake Ones During a Test Run

The AI model showed what it was capable of (Image via Google)
The AI model showed what it was capable of (Image via Google)

Artificial intelligence can be used as a crutch to make work more convenient. However, the lack of proper oversight could turn it into a problem. Heather Adkins was transparent about the outbreak of Google’s AI model and this incident will be a learning experience for the company moving forward.

“These events highlight the importance of training powerful AI models to act responsibly,”

Also Read: “You Really Need To Start Executing the Women Doing This”: Controversial Streamer Comments on Reports About Newborn Being Found Alive After Being Tossed Into Trash Can



Source link

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top