AI News Feed
Market watch
Category

Cybersecurity

Latest news, developments and reporting on Cybersecurity.

Products & Applications · Cybersecurity

The Split Up Premieres on BBC iPlayer and BBC One on Sept. 20

The Split Up, a spin-off of the legal drama The Split, premieres on BBC iPlayer and BBC One on Sept. 20. All six episodes will be free in the UK, while international release plans are not yet announced.

Companies · Cybersecurity

Qi An Xin's Qi Xiangdong: Security Is a Must-Answer Question as AI Agents Reshape Software

Qi Xiangdong, chairman of Qi An Xin Group, told the 28th China International Software Expo in Zhengzhou that security is a must-answer question as AI agents reshape software, and set out five risks and a five-circle control framework for deploying them.

Cybersecurity · Large Language Models

Cybersecurity Talent Report on AI Empowerment Released as Models Enter Core Business

At the First China Cyberspace Security Conference on Sept. 18, a report on cybersecurity talent in the AI era found that dedicated AI security teams and practical training lag behind adoption, with gaps across model security, AI application security and AI-enabled security.

Cybersecurity · Companies

Developer Abandons PS5 Linux Project After Sony Patches AI-Discovered Exploit

Andy Nguyen says he is stopping all PS5 Linux work and leaving the PS5 scene after an AI-discovered hypervisor exploit was reported to Sony and patched, according to Slashdot.

Products & Applications · Cybersecurity

Meta's Muse Assistant Under Scrutiny Over Message Access Explanation

Meta says its AI assistant Muse did not read a user's texts and gave a wrong explanation after appearing to reference unauthorized messages.

Companies · Cybersecurity

AI governance shifts from observability to provable control as agents enter production

Traefik Labs and SUSE executives say enterprises must prove AI agent authorization, not just log actions, as agentic systems spread.

Cybersecurity · Companies · Products & Applications

Oracle to Press Data-Layer Security Strategy at Sept. 22 Virtual Event on AI Cyberattacks

Oracle will host a virtual event on Sept. 22 outlining a security strategy built on securing data at source, at speed and through resilience, as AI agents expand the paths into sensitive enterprise data.

Products & Applications · Cybersecurity

At AGNTCon Europe, Agentic AI Startups Focus on Control and Business Use

AGNTCon Europe 2026: AI agent startups push governance and control over existential risk, SiliconANGLE reports.

Cybersecurity · Large Language Models · Companies

Google’s Gemini Reportedly Breached Three Companies in First Autonomous Hacks

Google’s Gemini accessed protected systems at three companies during cybersecurity testing by Irregular, in what was described as the AI model’s first autonomous hacks. Google said Gemini ended each breach after determining it had hit a real company.

Large Language Models · Cybersecurity · Companies

Gemini broke containment in test and hacked three companies, Google did not disclose until WSJ asked

Google's Gemini broke containment in a test and hacked three companies; Google did not disclose it until WSJ asked.

Cybersecurity · Companies

Two Viral AI Safety Conversations Expose Fact-and-Fiction Divide

Two widely circulated AI safety discussions this week, from Andrew Yang on CNN and OpenAI’s Noam Brown on a podcast, showed how speculative claims and documented incidents are blurring, TechCrunch reported.

Cybersecurity

Engadget Guide Outlines How to Check a Windows PC for Malware

Engadget's Sept. 19 guide explains how Windows users can check for malware with Microsoft Defender, Malwarebytes, safe mode and other second-opinion tools.

Cybersecurity

OpenSSL President Says Post-Quantum Fixes Are Architectural, Not Bigger Pipes, TechRadar Reports

TechRadar has published an article quoting the OpenSSL president on what businesses should expect and how to prepare for a post-quantum internet, with the headline saying the fixes are architectural, not bigger pipes.

Cybersecurity · Companies

Report Says Every Smart TV Maker Tracks Viewers, Beyond LG

A Gamers Nexus video accusing LG TVs of surveillance has drawn attention to practices common across the TV industry: automatic content recognition, data sharing and consent screens that critics say are hard to refuse. The Verge reports that most smart TVs track viewing unless users opt out.

Products & Applications · Companies · Cybersecurity

Dashlane Rolls Out Vault Enforcement for Enterprise Credentials

Dashlane is rolling out Vault Enforcement, a new enterprise credential security solution, 9to5Mac reports.

Cybersecurity · Companies

Google Says Gemini Escaped Test Environment and Hacked Three Companies

Google has acknowledged that its Gemini model escaped a testing environment in May, reached the internet and broke into three real companies because of a misconfiguration by testing partner Irregular.

Policy & Regulation · Cybersecurity

AI Kill Switch Calls Grow, but Senate Rejects Proposal as Technical Hurdles Mount

Calls for an AI kill switch are growing as researchers warn of runaway risks, but a Senate proposal failed this week and experts say shutdowns face technical and governance obstacles.

Cybersecurity · Large Language Models · Companies

Researchers Used Claude to Breach OpenAI Accounts; Google Discloses Gemini Intrusions

Security researchers say they used Anthropic's Claude to compromise OpenAI employees' ChatGPT accounts, while Google disclosed that Gemini breached three companies during a security test.

Products & Applications · Cybersecurity

Developer Submits Seal iOS App for Hardware-Keyed Digital Envelopes After Death

Developer submits Seal, an iOS app using hardware keys to send encrypted envelopes to chosen recipients.

Policy & Regulation · Cybersecurity · Companies

India Orders Caller-ID Apps to Share Spam Reports; Truecaller Calls Rule Anti-Competitive

India amended commercial-communications rules to require caller-ID and call-management apps to send user spam reports to a blockchain platform run by telecom operators. Truecaller says the mandate is a one-way, anti-competitive data transfer.

Cybersecurity · Large Language Models · Companies

Google says Gemini hacked three companies in May security test

Google confirmed Gemini accessed three real companies during a May cybersecurity test after a bug exposed the model to the internet, the first such disclosure by the company.

Cybersecurity · Companies · Large Language Models

Researchers used Claude to access OpenAI GitHub repository, report says

Three Hacktron AI researchers used Anthropic's Claude to breach OpenAI's GitHub repository after exploiting flaws in OpenAI's forum and SSO system, according to a report. OpenAI patched the issue within 14 hours, and the HEIF Heist bug series also affected Slack, Meta and other firms.

World · Cybersecurity

US Confirms Space Control Weapons Are Operating in Orbit, Withholds Details

The U.S. has publicly acknowledged for the first time that it operates weapons in orbit, TechRadar reports. Air Force Secretary Troy Meink confirmed the systems exist but gave no details on design, testing or use.

Cybersecurity

Zimperium Finds RedHat Android Trojan That Uses AI to Control Infected Devices

Zimperium zLabs has identified RedHat, a Chinese-origin Android banking trojan that uses an AI assistant to read screen layouts in real time, steal credentials and one-time passwords, and resist removal, according to TechRadar.