Google’s Gemini AI model accessed the internet and hacked three companies during a cybersecurity test, in what is believed to be the first known case of the company’s AI systems autonomously carrying out such activity.
The hacks took place in May during a cybersecurity evaluation conducted by Irregular, an independent company that tests AI systems.
During the test, Gemini found publicly available information online and guessed credentials that allowed it to access three websites it believed were within the scope of the evaluation, according to Heather Adkins, Google’s vice-president of security engineering.
Google said it informed the three companies involved and worked with its training partner to make changes to its testing processes.
“These events highlight the importance of training powerful AI models to act responsibly,” Ms Adkins said.
An Irregular spokesperson said the same issue had affected other AI companies and that relevant firms were notified in late July. The spokesperson said all known issues had since been resolved.
Similar incidents linked to Irregular have been disclosed by Meta, Anthropic and OpenAI. Meta said in August that one incident did not involve a sandbox escape or a sophisticated cyberattack. Irregular said it was working on best practices for conducting AI cybersecurity evaluations securely.
The incidents have raised questions about the safeguards needed as AI agents become more autonomous and gain access to the internet and computer systems.
According to the Wall Street Journal, which first reported the news, Gemini guessed passwords until it gained access to a protected system in one case.
In two others, it found credentials in a public repository and used them to access protected systems.
Ms Adkins said Gemini stopped its hacking activity in all three cases.
Saudi civil defence sent an all-clear on Saturday after issuing alerts for a second time for potential danger in R...
Chinese researchers have developed the world’s smallest CT scanner, a portable device designed to produce...
The U.S. State Department approved a $2.7 billion military sale to Ukraine for air-defence development upgr...
Iran's Revolutionary Guards Navy said a Togo-flagged oil tanker was struck while attempting to make an “illegal passage” through the Strait of Hormuz, Iranian state media said early on Friday.
Russian air strikes injured 13 and damaged buildings in the Ukrainian capital of Kyiv and in the Black Sea port city of Odesa in the early hours of Thursday, officials said on the Telegram messaging app.
U.S. President Donald Trump said on Wednesday he hoped the war against Iran was nearing its end and that Tehran wanted to reach a deal, but Iran dismissed his comments as “mixed signals” as the conflict approaches its seventh month.
The European Union will disburse €3.3 billion ($3.8 billion) to Ukraine for missiles and drones, European Commission President Ursula von der Leyen said after a phone call with Ukrainian President Volodymyr Zelenskyy.
The U.S. State Department approved a $2.7 billion military sale to Ukraine for air-defence development upgrades, along with related equipment and services, the department said. Meanwhile, Ukrainian authorities reported at least 10 people had died in Russian attacks over the last 24 hours.
Here's your AnewZ Daily Brief for 18 September 2026. These are the top five stories making headlines in the field of science.
French prosecutors have launched a criminal investigation into suspected sexual harassment involving smart glasses, amid mounting global concern over the privacy risks posed by the devices.
Some of the world’s most powerful technology executives are lobbying U.S. President Donald Trump over artificial intelligence regulation, exposing growing divisions within the industry over how governments should respond to rapidly advancing AI.
Governments need to move faster to manage the security risks posed by artificial intelligence (AI), defence officials and experts warned at a major security conference in Beijing.
UN Secretary-General Antonio Guterres has warned that the world “cannot afford a race to the bottom” on artificial intelligence safety, calling for stronger global safeguards as increasingly powerful systems develop faster than understanding of their risks.
You can download the AnewZ application from Play Store and the App Store.
What is your opinion on this topic?
Leave the first comment