News for today:
- [Janusnet] Accord announced. Janusnet announced a new product – Janusnet Accord – for enterprise metadata interoperability between systems, networks, and allied partners; Accord ensures that data security requirements remain consistent. Janusnet Accord addresses [the] interoperability challenge [across coalition operations] by validating, translating, and standardizing security metadata across legacy, existing and new architectures. This enables downstream systems to enforce policy decisions based on authoritative, trusted metadata. Janusnet
- [BeyondTrust] Forcepoint integration for data blast radius mapping and remediation. BeyondTrust and Forcepoint announced a new product integration – with identity risk signals from BeyondTrust triggering Forcepoint to show the sensitive data at stake. Customers now get the data context behind each identity risk, along with the controls to act on it. BeyondTrust Identity Security Insights surfaces risky identities based on behavior, privilege, compromised credentials and relationships among them. Forcepoint Data Security Posture Management (DSPM), part of the company’s AI data security solutions, maps that risk to the confidential information within its permissions and provides controls to restrict that access. BeyondTrust or Forcepoint (basically the same press release at both places)
- [LastPass] New Plus offering with VPN. LastPass released a new plan – LastPass Plus – combining its password manager with a new consumer-oriented VPN. With LastPass Plus, customers can extend trusted credential protection to their browsing activity, all without deploying a new account, application, or vendor. Many LastPass consumers already use a third-party VPN, which means they’re juggling multiple security applications, subscriptions, and logins to help keep both their credentials and browsing activity safe. LastPass VPN can remove those additional costs and complexity barriers by offering customers a unified privacy and security experience inside the same LastPass ecosystem they already trust. Available immediately, but for Windows only initially. LastPass
- [AppViewX] Visibility and a kill switch. AppViewX expanded the capabilities of its Agent Identity Security solution with shadow AI agent discovery, a new MCP gateway, and an agent kill switch. On the first: AppViewX discovers sanctioned and unsanctioned agents across the enterprise, with new capabilities to discover browser-based and short-lived, script-based shadow agents. The AI Bill of Materials (AIBOM), which gives security teams a centralized view of every agent’s models, MCP tools, credentials, and identities, now extends to the skills agents can access, so that teams can assess what these skills enable and flag unsafe capabilities. Unlike alternative solutions that rely on a single detection method, AppViewX combines its own new lightweight endpoint Guardian Agent with API integrations across EDR, SaaS, and cloud platforms, giving security teams confidence that no shadow agent goes undetected. AppViewX
- [Aembit] Access controls for personal AI agents. Aembit said it’s now supporting identity controls for personal AI agents within enterprise environments, to control data access and change requests initiated by said agents. As personal agents, such as Meta Muse and OpenAI Dots, move into workplace use, they can gain access to email, cloud applications, code repositories, internal tools, and other enterprise services through employee credentials and delegated permissions. That raises the risk of sensitive data exposure, unauthorized changes, and activity that is difficult to distinguish from the employee’s own actions. Those credentials may also be stored in the personal agent platform, outside the security team’s direct control. Aembit
- [AegisAI] Proteus for email security. AegisAI Labs released its first internally trained email security model – for detecting attacks, many of which are now written by machines, not people. Investigating every attack is what makes the difference in detection. On a benchmark of 3,020 human-reviewed phishing attacks drawn from one week of production traffic, Proteus identified more than 99% of attacks. A leading closed-source general-purpose model identified 81.3%, and a widely used open-weight model 80.0%. Each general model let through roughly thirty times as many attacks as Proteus. Proteus was also faster. Against the closed-source model it returned its verdict 7.2x sooner at the median, and 26x sooner in the slowest 1 in 20 cases. AegisAI
- [Radware] Service center in Thailand. Radware expanded its global cloud security network with a new service center in Thailand. The new center … is designed to provide organizations in Thailand with local application and API protection, with the goal of helping reduce latency and supporting compliance with data and regulatory requirements. The expansion builds on Radware’s growing business with government and enterprise customers in Thailand and strengthens the company’s ability to meet increasing demand for locally delivered cloud security services. It also comes as Thailand strengthens cybersecurity requirements for cloud environments and cyberattack activity rises across Asia Pacific. Radware