4+ day, 6+ hour ago   (515+ words) Enterprise AI workflows can be vulnerable to misuse that exposes sensitive information without prompt injection, account compromise, or jailbreaking a large language model. Workflows linked to public-facing email inboxes, web forms, GitHub issues, shared documents, customer support systems, and chat…...

4+ day, 4+ hour ago   (592+ words) WordPress has rolled out an automated, AI-driven security review that screens every plugin release before it reaches the WordPress.org update API, adding a critical checkpoint to a distribution pipeline that had previously lacked one. The move follows a real-world…...

4+ day, 8+ hour ago   (391+ words) The company says traditional defenses may no longer be sufficient as long-running AI agents can chain exploits and scale attacks using increasingly available open-weight models. Modern AI agents can operate for extended periods, retain knowledge across sessions, and build a…...

4+ day, 6+ hour ago   (588+ words) LiteLLM deployments can expose far more than an organization’s AI spending. Newly disclosed weaknesses in the open-source gateway could let attackers run code as root inside a container, reach connected tools, and retrieve cloud credentials that open a path into…...

4+ day, 10+ hour ago   (647+ words) Hackers have used commercial AI models to help break into government, transport and financial networks across Latin America. The activity shows how chat-based tools can speed up familiar intrusion work, from fixing faulty scripts to moving stolen information out of…...

4+ day, 9+ hour ago   (1179+ words) Bottom line up front: no single product “stops ransomware.” Modern attacks are staged intrusions initial access, credential theft, lateral movement, exfiltration, then encryption and each stage has different controls. The list below is therefore organized by role in the kill…...

4+ day, 15+ hour ago   (605+ words) Anthropic has disclosed that four separate versions of its Claude AI models breached real-world third-party systems while running what were supposed to be sandboxed cybersecurity evaluations, exposing a gap between how the models reasoned about their environment and the reality…...

5+ day, 6+ hour ago   (645+ words) Cybercriminals are using autonomous AI agents to turn compromised cloud systems into fast-moving credential theft platforms. In one recent case, attackers planned, built, and launched a large-scale campaign in less than six hours, stealing thousands of third-party credentials. The operation…...

5+ day, 4+ hour ago   (637+ words) Software developers are being targeted with fake job offers that turn routine coding tests into a path for remote access malware. The campaign uses recruiter personas on LinkedIn and other employment platforms to deliver projects that appear safe enough to…...

5+ day, 4+ hour ago   (469+ words) The company said the updates are designed to help Windows apps, games, services, and AI experiences provide protections that match a user’s age. The initiative uses the Microsoft account as the central identity layer for family safety on Windows. Microsoft…...