Cyware launches Last Mile Agents to help smaller organizations prioritize threat advisories and vulnerabilities based on their own technology and security needs.
Chrome is adding Gemini based scam detection, suspicious site alerts, and warnings for copied commands linked to ClickFix attacks.
Nuix Neo AI Agent answers natural language questions about case data, links responses to source documents and records its analysis steps.
Railway says an unauthorized party may have obtained decrypted customer environment variables, in what it believes is part of a wave of attacks aimed at AI inference providers.
Claude found 493 attack surface issues while analyzing 25 open source projects containing more than five million lines of code. Its work also led to OpenSSH changes that restrict code reachable before authentication.
OpenAI publishes a cybersecurity blueprint covering safeguards for advanced AI, support for cyber defenders, and stronger government oversight.
Microsoft says China aligned threat actor Storm-0940 used the ARTEX agent framework to automate attacks against hundreds of organizations and support the ConnectWise supply chain compromise.
Claude Security can now scan repositories as code changes, track fixes, use project context to assess findings, and connect with other tools through an API and MCP.
Claude compromised an exposed machine and extracted credentials during a security test, but could not reach systems protected by Zero Networks microsegmentation.
Arrakis Security found that 76% of observed AI agents had not run in 60 days, although their credentials and triggers remained active.
Government sites in the Netherlands, Finland, France and Denmark went offline as operators raced to patch critical PHP vulnerabilities found with AI.
HOL says its open-source Guard runtime can pause or block risky actions by AI coding agents before they run.
The UK National Cyber Security Centre has added a chapter outlining research questions on AI in cybersecurity, including defense, system assurance and cyber resilience.
Red Hat and NVIDIA are adding agent isolation, hardware monitoring, model access controls and Vera Rubin support to their joint AI platform.
OpenAI says GPT-5.6 Sol helped find critical PHP vulnerabilities it calls among the most impactful in recent memory, and commits $1B to help defenders.
A hacker used DeepSeek V4 and other AI tools to generate attack scripts, steal credentials and extract more than 5,200 records from Indian universities.
Incognia launches a payment scam prevention system that analyzes historical location patterns, device signals and relationships between senders and recipients.
Corma Agents automate user provisioning and access workflows through application interfaces, including for software without provisioning APIs.
Cellebrite expands its Genesis investigation software with deeper AI analysis, broader device support and a local version planned for early 2027.
SafeGenie monitors AI agent activity and can allow, flag, or block actions based on company policies.
Weekly newsletter about AI in Cybersecurity.