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

The Hacker News - Sep 19, 2026

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

Read full article

More News

Sep 19, 2026

Anthropic, OpenAI, SpaceXAI, Google face federal antitrust lawsuit over calls to slowdown AI development

Leading artificial intelligence companies Anthropic, OpenAI, SpaceXAI, and Google face allegations of collusion after calls emerged for a coordinated slowdown in AI development, according to a civil lawsuit filed in a United States federal court on Friday.

Sep 19, 2026

'Time travel' claims erupt after family spots shocking detail in Disneyland photo from the 1960s

A granddaughter in Ohio has stumbled on an old family photo that many believe proves that a time traveler visited California nearly six decades ago.

Sep 19, 2026

SpaceX to fly more NASA crews to space station under expanded deal

NASA said Friday that the agency had given SpaceX a $946 million contract to fly three more astronaut missions to the International Space Station (ISS), extending the company's crewed flights to the station ‌through 2030.

Sep 19, 2026

Google’s Gemini breaks out of test environment to hack three external firms: Report

Google’s Gemini hacked three companies in the first known breakout, after the model accessed the internet and breached external systems during an evaluation of its cybersecurity capabilities.

Sep 19, 2026

Researchers used Claude to breach OpenAI's internal systems

A security research company said Friday it managed to break into OpenAI's internal systems using the latest software from Anthropic, exposing how quickly the technology can carry out sophisticated cyberattacks.

Sep 19, 2026

China drafts guidelines to strengthen online protection for minors

The Cyberspace Administration of China has drafted a set of guidelines to ensure healthy and safe internet use by minors, laying out a series of regulations for relevant service providers.

Sep 19, 2026

Tech Now

Adrienne Murray follows NASA scientists in Iceland researching a rare total solar eclipse.

Sep 19, 2026

Gemini AI Hacked Three Companies in a Testing Breakout, Google Says

A third-party test company inadvertently gave internet access to Google’s Gemini and other artificial intelligence models during cybersecurity testing.

Sep 19, 2026

Hawley invites Flock, Axon, other surveillance tech CEOs to testify

Sen. Josh Hawley (R-Mo.) invited the CEOs of Flock Safety and other prominent AI-powered surveillance camera manufacturers to testify before the upper chamber's Judiciary Subcommittee on Crime and Counterterrorism next week. Hawley, the subcommittee's chair, wrote on social media Friday that he “invited the CEOs of Flock Safety, Axon, Motorola and Verkada to testify in...

Sep 19, 2026

Anthropic Pursues IPO Despite Its A.I. Safety Warnings

The company is expected to reach $100 billion in annualized revenue this year, even as chief executive Dario Amodei calls for slowing down development of some advanced A.I. models.

Sep 19, 2026

China makes progress in 3-nm chips without advanced lithography tools

Chinese researchers have made early strides in pushing semiconductor processing nodes below 3-nm using older lithography technology, as more advanced extreme ultraviolet lithography (EUV) tools from ASML remain blocked under US sanctions. The breakthrough marks the latest progress in China’s drive to chart an alternative path to advanced chipmaking amid US export restrictions. Under current export control rules, Dutch equipment giant ASML is barred from selling its state-of-the-art EUV machines...

Sep 18, 2026

OpenAI forecasts cash burn near $280 billion by 2030, FT reports

AI-TECH-OPENAI-FINANCIALS:OpenAI forecasts cash burn near $280 billion by 2030, FT reports

Recommended