Latest news, developments and reporting on Cybersecurity.
Ubuntu 26.10 (Stonking Stingray) is available for download, with an upcoming Linux 7.3 kernel, GNOME 51, TPM-backed full disk encryption and Rust coreutils.
Citizen Lab reports the first public case of zero-click spyware delivered via targeted ads, affecting Al Jazeera devices.
Group-IB researchers have analyzed KGB RAT, a malware-as-a-service remote access trojan that steals data and cryptocurrency while using jump scares, screamers and fake blue screens. It is sold from $150 per month to $800 annually and has been marketed as legitimate software.
Citizen Lab says mercenary hackers used targeted ads to deliver zero-click exploits that infected 45 Al Jazeera devices in 2024 and 2025, likely the first confirmed case of ad-delivered zero-click mercenary spyware.
FTC subpoenas OpenAI and Anthropic for documents on safety incidents and possible consumer protection violations.
IBM ties intelligent operations to agent safety and business value, Tyler Lynch says ahead of IBM TechXchange 2026.
Meta’s Muse AI assistant faces privacy concerns for collecting data about users’ contacts and friends.
Bank of America hired 1,000 contractors and spent more than $100 million fixing software flaws uncovered by Anthropic's Mythos, CEO Brian Moynihan said. Other major banks are also spending heavily as advanced AI security tools expose vulnerabilities faster than teams can patch them.
AI governance startup nFOX Lab launched a platform to manage privately owned large language models and said it plans to open-source Blockhouse, a Linux-level sandbox that removes AI workload privileges by default. The company says the tools are designed to help enterprises run open models on their own infrastructure while limiting security risks.
Microsoft's cumulative updates KB5127646 and KB5127648 fix Always On VPN failures on Windows 11 24H2 and newer, but some users report the patches have broken enterprise Wi-Fi authentication.
Microsoft, Proofpoint and Japanese law enforcement dismantled YAM Store, CoGUI and Iron Man, a Chinese-language cybercrime marketplace that sold phishing tools, accounts and infrastructure. Microsoft says the network facilitated more than a billion phishing messages this year before domains were seized and over 20,000 accounts disabled.
Benn Jordan testified before a Senate subcommittee on Flock Safety's license plate readers as cities cancel contracts.
Rockstar says 2022 GTA 6 leaks were 'weirdly invasive' and led to a five-day in-office policy it may review after launch.
The CNBC Investing Club said Wall Street has grown more divided as AI-related selling hit tech, with Micron the week's biggest loser and Cardinal Health a defensive winner. Over the five weeks through Wednesday, CrowdStrike and Meta led its 32-stock portfolio, while BNY, Goldman Sachs and FedEx Freight lagged.
Citizen Lab and Confiant say Discover-Tech worked with spyware companies to deliver zero-click attacks through targeted ads, including Candiru-linked spyware that hit at least 45 Al Jazeera devices in 2024 and 2025. Google, Beeswax and IAB Europe have taken action.
Private Internet Access added 10 server locations across the Americas, the Middle East, Africa and Australia, following 15 additions in September. The company says its network now covers 170 locations in 104 countries, though independent tracking points to 176 in 112.
Descope and ServiceNow unveiled tools to govern employee-created AI agents through identity, access and policy controls.
TechRadar published an Oct. 15, 2026, article on AI fakes to warn friends and family; it is by contributor Becca.
TechRadar says Private Internet Access's Halloween Sale is its VPN Deal of the Week, offering 26 months for under $50.
Ledger has announced Atom, an Android-based, data-only hardware wallet that looks like a phone but works alongside a smartphone, with reservations open now and availability set for summer 2027. The launch comes as Ledger pauses some sales after a supply-chain tampering incident.
Bank of America brought in 1,000 outside contractors to patch software vulnerabilities exposed by advanced AI models including Anthropic's Mythos, adding hundreds of millions of dollars in third-quarter costs, CEO Brian Moynihan said.
The Verge reports that Energize Lab’s Eiliko robot toy left 13 million conversations and detailed user profiles on unprotected Chinese servers. A researcher could access about 68,000 active robots, including more than 31,000 child profiles in the United States, and the company did not meaningfully respond to questions.
Citizen Lab says online ad networks were abused to deliver zero-click spyware, including attacks on Al Jazeera devices.
Proofpoint added a Microsoft 365 tenant risk dashboard, Teams security training and new AI features to its MSP Platform suite.