Cybersecurity
Hacker Used AI Agents to Breach More Than 100 Sites, Including John Deere and SharkNinja
Norforge says a hacker used AI agents to breach over 100 sites; Reuters names John Deere, SharkNinja and Cummins.
Latest news, developments and reporting on Cybersecurity.
Norforge says a hacker used AI agents to breach over 100 sites; Reuters names John Deere, SharkNinja and Cummins.
Axonius launched AI Workforce virtual workers and made its AI Companion and MCP Server generally available, while Fortanix extended Confidential AI to Kubernetes through CNCF Confidential Containers. Both aim to control AI-driven remediation and protect data during inference.
Burger King Japan says unauthorized access to its app member system between Oct. 2 and Oct. 7 exposed 11.18 million customer records, including names and order histories, but not payment data.
A director at Cloudhouse warns that unmanaged IT change can cause outages and cyber attacks, and recommends three steps to control it, according to a TechRadar article.
A Tech Transparency Project investigation found 13-year-old test accounts could download 135 apps that strip clothing or swap faces onto nude bodies, despite parental controls and store bans.
AI writes half of committed code in adopting firms, but only 56% of tested tasks pass security checks, TechRadar reports.
Hollys said 2.76 million pieces of customer data were leaked in a cyberattack between Sept. 3 and Oct. 7, following data breaches at 16 South Korean financial institutions.
Federal prosecutors say Abdulrahman Alfozan, accused of hiding a recorder in an OpenAI conference room, was in frequent WhatsApp contact with a member of the Saudi royal family, including four days before his arrest.
South Korea's FSS chief vows stronger cybersecurity after breaches exposed about 100,000 personal data cases.
South Korea’s FSS chief pledged stronger cybersecurity and probes after attacks exposed about 100,000 data cases.
Savannah officer fired for giving Flock login to ICE agent husband, who ran 983 searches tied to 97 plates.
Yelp, Delta and banks block AI agents as Amazon restricts Meta's Muse, raising questions about agent usefulness.
OpenAI's Alfozan arrested by FBI over alleged conference-room recorder; bail sought.
Fake WatchGuard VPN page stole admin credentials while serving the vendor's signed client, Huntress says.
TechRadar lists free Chargers-Chiefs options, though its TVNZ location differs between headline and body.
Palm OS developer releases 1.5 GB archive and rediscovers a 2001 virus he wrote, keeping working code private.
Austrian MEP Lena Schilling, targeted by nude deepfakes, urges enforceable EU curbs after a report found 50 MEPs affected.
Financial Supervisory Service data reported by Yonhap News show that South Korea's five major banks confirmed 8,760 fraud-flagged voice phishing transactions in 2025, more than double the 2024 total.
A federal judge has paused Utah's SB 73, which required adult sites to block VPN users, ruling it likely violates the Constitution.
Rental cars used by Norway's army in Oslo stored dashcam footage of soldiers and gate codes, TechRadar reports.
ETS breach exposed names, birth dates and addresses of millions of California public school students.
Samsung Galaxy phones with One UI 6.0 or later include Auto Blocker, a default security feature that blocks risky app installs, USB commands and message images. Engadget reports that the tool also complicates sideloading, though Samsung recommends keeping it on.
Ad-delivered zero-click spyware infected at least 45 Al Jazeera devices, researchers say, linking the campaign to Candiru and ad-tech firm Discover-Tech.
Le Monde reported on 17 October 2026 that a cyberattack hit a French supermarket's refrigeration systems, presenting the incident as a warning to the retail industry about cybersecurity weaknesses.