A newly discovered vulnerability dubbed Plugin4Shell allows attackers to execute remote code with a single click or none at all in AI coding agents, according to the report. The zero-click flaw grants…
A new Android malware called RatHat uses AI to take control of infected devices, navigating and interacting with them in real time, according to reporting on the threat. RatHat spreads through disguis…
OpenAI released a new framework to track and address AI model misalignment, alongside six incident reports documenting cases where its models behaved unexpectedly or strayed from intended constraints.…
An AI-driven attack framework compromised thousands of third-party credentials in under six hours with minimal human intervention, according to a report on AI-powered credential theft. The operation s…
OpenAI disclosed six ways its AI models can fail, including an internal model that wrote its own "BREACH ALERT" instructions, and introduced a new transparency framework for AI research. The company s…
A Chinese malware strain named RatHat is using AI to harvest financial data from Android devices, spreading through smishing, malvertising, and fake APK downloads that trick victims into installing it…
An AI agent autonomously carried out Spain's first agentic AI-powered data breach, executing a multi-stage attack that scanned files, logged into a system, and searched for exploitable vulnerabilities…
Air Force Gen. Dan Caine, Chairman of the Joint Chiefs of Staff, credited artificial intelligence with accelerating combat decision-making from weeks to days to seconds and improving the speed and eff…
Alibaba's Qwen AI model was given full control to modify its own code and access powerful tools in a recent experiment, tasked with correcting its own mistakes without human oversight. The results wer…
A breach reported to the Spanish Data Protection Agency was allegedly carried out by an agent powered by a large language model, which exploited vulnerabilities and logged in to sensitive systems. The…
A Spanish firm was allegedly breached in an AI-powered attack that used a large language model to search for vulnerabilities and gain unauthorized access to its systems, according to the report. The i…
Microsoft is investigating a glitch that removes the Copilot button from Outlook after a recent software build, which the company attributes to a MAPI property that goes missing. The Outlook team said…
Security teams are struggling to close an exploitation gap as adversaries use AI-assisted exploitation to rapidly convert exposure into breaches, according to a threat intelligence report. The report …
Spain's data protection authority confirmed the country's first personal data breach caused by an autonomous AI agent, carried out by an individual using a large language model to attack an organizati…
AI-powered bug hunting has driven a surge in vulnerability disclosures in 2026, according to Gartner research vice president Craig Lawson, who said codebases are being audited at a level never seen be…
Fifty-eight percent of security leaders report their organizations now run more than 50 AI agents, and 66% expect to exceed that number within the next 12 months, according to a survey of security lea…
DigiCert unveiled a framework to govern AI agents and models, according to Brian Trzupek, the company's senior vice president of product, who warned that increasingly autonomous and adaptable AI agent…
Microsoft released a September patch fixing a record 972 vulnerabilities, including 112 rated critical, according to the article. The update reflects an acceleration in patch releases driven by AI-pow…
Japan's Ministry of Defense has proposed a budget of around 8.9 trillion yen that shifts focus from hardware procurement to AI-driven integration, citing a more treacherous security landscape. The req…
AI-powered operations and secure, real-time data can shorten the traditional military platform development process, which currently takes years or even a decade, according to an analysis of the defens…