The barrier to entry for local AI has officially collapsed. In 2026, running a highly capable Large Language Model (LLM) no longer requires a massive tower with multiple discrete GPUs pulling 1000 watts. If you are building agentic frameworks, running local log analysis for Identity Threat Detection and Response (ITDR), or just want an uncensored…
KPMG and Microsoft scale trusted, enterprise AI agents globally through deployment of Agent 365 and Copilot. AI and Data coverage from iTWire.
Rubrik Launches Rubrik Agent Cloud for Anthropic’s Claude Code. Cloud coverage from iTWire.
Drata has introduced AI Agent Governance, a new security category focused on managing the risks and oversight requirements of AI agents, while extending its trust platform to support enterprise adoption of autonomous AI systems. While McKinsey finds 57% of business leaders cite governance friction as the top blocker to deploying more AI, this move is…
Rubrik introduced Rubrik AI, an agent-first interface for its Security Cloud and Agent Cloud, allowing customers to define business outcomes that the software executes by reasoning over data, identities, and deployed agents.
Phishing simulation on an OpenClaw email agent with various configuration profiles showed that it was susceptible to tactics commonly used to compromise human users. […]
The new integration evaluates every access request made by a Copilot agent in real time, providing a decision before the action is executed.
Ox Security field CTO, Boaz Barzel, makes the case for vibe security to tackle AI agent coding risks
The new agent is part of the company’s effort to broaden beyond its core consumer business as it spends aggressively on artificial intelligence.
Microsoft has developed a new AI agent that can run autonomously around the clock to complete tasks across Microsoft 365 applications. Microsoft Scout, unveiled at the company’s Build event Tuesday, is a new type of always-on agent based on the OpenClaw agent framework that Microsoft calls “autopilots.” These act on a user’s behalf with their…
Noma has announced the launch of Noma Agent Access Control, which helps security teams discover, govern, and enforce access policies for AI agents and Model Context Protocol (MCP) servers throughout the enterprise. AI agents and MCP servers have proliferated across developer environments faster than existing governance frameworks were designed to handle. In less than 12…
An unknown threat actor has been observed using a large language model (LLM) agent to conduct post-compromise actions after obtaining initial access following the exploitation of a publicly-accessible Marimo network using a recently disclosed vulnerability. “The attacker compromised an internet-reachable Marimo notebook via CVE-2026-39987, extracted two cloud credentials from the compromised
Claroty has launched Claroty Claire, a CPS-native AI security agent designed to help organizations defend mission-critical infrastructure. Claire is powered by a CPS language model trained on more than a decade of industry expertise and CPS-related data. The launch expands organizations’ capabilities for supporting the safety, uptime, and availability of cyber-physical systems. Defending a rapidly…
Ketch has unveiled its vision for agentic privacy with the Ketch Agent Network, a multi-agent orchestration layer for enterprise privacy programs. The platform is designed to continuously reason across legal obligations, internal policies, and operational realities within a unified AI-driven system. Privacy teams today are accountable for more than ever: global regulations, AI governance mandates,…
Checksum has launched its Continuous Quality Agent, an autonomous system that runs nightly against deployed applications and automatically heals broken tests without waiting for an engineer to open a dashboard or write a prompt. AI coding has changed the constraint in software development. Teams can now ship far more code than before, but every PR…
The updated platform introduces Xage Agent Sentry and Xage Resource Gateway, which aim to secure AI agents and the resources they access at multiple levels, including network interactions, local events, and operating system calls.
Google has introduced Agent Executor, an open source runtime aimed at helping enterprises run AI agents more reliably at scale, as attention shifts from building agent prototypes to managing the operational challenges of putting them into production. To address those production-related challenges, the runtime, according to the company, comes with capabilities that are geared towards…
Kore.ai has launched the new-generation Kore.ai Agent Platform Artemis edition, the AI-programmable, AI-native foundation that builds, governs, and optimizes the agents, systems, and workflows running across the enterprise. The platform launches initially on Microsoft Azure, with broader cloud availability to follow. The new-generation Kore.ai Agent Platform enables enterprises to deploy production-ready multiagent AI systems in…
Google is expanding the role of its CodeMender security agent from autonomous vulnerability remediation toward a larger agentic development ecosystem, signalling a broader push toward AI-driven AppSec. Months after introducing CodeMender, an AI-powered agent designed to autonomously identify and patch software vulnerabilities, Google is now integrating the technology into its expanding Agent Platform strategy unveiled…
This is Part 2 of a 2-part series. Read Part 1: Your AI Agent Doesn’t Care About Your Controls If AI agents change how execution happens, they also expose a fundamental limitation in how most security controls operate. Many control models assume there is sufficient time to detect, assess, and respond to events before they result in…
Microsoft has open-sourced two tools aimed at bringing security discipline to AI agent development: Clarity, a structured design review tool, and RAMPART, a continuous testing framework. The release comes from Microsoft’s AI Red Team, the company’s internal unit that stress-tests its own AI systems, and both tools have been used internally before being open-sourced. RAMPART:…
AI agent projects are proliferating throughout the enterprise, and those AI agent identities require management, security, and governance. New Omdia research shows the AI agent identity budget dynamics are very different than traditional IAM projects.
AI agent projects are proliferating throughout the enterprise, and those AI agent identities require management, security, and governance. New Omdia research shows the AI agent identity budget dynamics are very different than traditional IAM projects.
Autonomous agent orchestration tool OpenClaw hit the scene last November and immediately went viral, but its dramatic flaws were exposed just as quickly. Still, it marked a pivotal step in the agentic AI era, and enterprises have been exploring ways to deploy fleets of autonomous agents safely and securely ever since. Automation Anywhere Tuesday rolled…
AI agent security starts with a simple fact: the more authority an agent has, the tighter its access…
The now patched vulnerabilities in the rapidly growing AI agent framework allow attackers to steal credentials, escalate privileges, and maintain persistence.
A critical vulnerability affecting certain configurations of the Exim open-source mail transfer agent could be exploited by an unauthenticated remote attacker to execute arbitrary code. […]
Zimperium launched Mobile App Response Agent, enabling security teams to respond faster than ever before to fraud and security threats. Leveraging Zimperium’s expertise in mobile security, Mobile App Response Agent is part of Zimperium’s Mobile App Protection Suite (MAPS), empowering SOC and fraud teams to assess attacks on their mobile app before they result in…
Dubai-founded OTT Cybersecurity LLC also unveils the Agent Trust Protocol (ATP), the first open cryptographic standard for AI agent identity, scope, and action verification — slated for IETF submission. OTT Cybersecurity LLC, the company behind Lyrie.ai, today announced two milestones that together position the company as foundational infrastructure for the agentic AI era: acceptance into…
Securonix announced the Securonix Threat Research Agent and ThreatWatch for ThreatQ, expanding how security teams research threats, validate exposure, and turn intelligence into documented action. Built on the ThreatQ platform and connected to Securonix security operations workflows, the new capabilities help teams generate role-specific intelligence, validate emerging threats against historical telemetry, and deliver explainable findings…
Extreme Networks has introduced Extreme Agent ONE, a new class of AI agents for enterprise networking. Moving beyond generic, prompt-based AI, Extreme Agent ONE runs on the Extreme AI stack purpose-built for enterprise environments, which combines advanced AI reasoning, live network context, and operational expertise to transform enterprise networks into systems that detect, decide, and…
ServiceNow has launched Autonomous Security & Risk to govern every AI agent, identity, and connected asset. Armis delivers continuous asset intelligence across code, IT, OT, IoT, and connected assets, while Veza provides fine-grained visibility, intelligence, and governance for both human and non-human identities. Security and risk crossed $1 billion in annual contract value (ACV) for…
An AI agent that revealed sensitive data without being asked. An agent that overruled its own guardrails. Another that sent credentials to an attacker via Telegram, because it forgot it wasn’t supposed to do so after a reset. It’s no secret that AI agents have huge potential, balanced by equally big risks. What’s becoming apparent,…
An AI agent that revealed sensitive data without being asked. An agent that overruled its own guardrails. Another that sent credentials to an attacker via Telegram, because it forgot it wasn’t supposed to do so after a reset. It’s no secret that AI agents have huge potential, balanced by equally big risks. What’s becoming apparent,…
The issue isn’t artificial intelligence, but rather an industry adding AI agent integrations into production environments before proper security testing.
A US agent claimed WhatsApp encryption is fake and Meta can access messages; the probe was abruptly shut, raising security concerns. A US agent claimed WhatsApp encryption is fake, alleging Meta accesses all unencrypted messages, but Commerce Department abruptly shut the probe, leaving leaders questioning if consumer apps are safe for sensitive business decisions. In…
PocketOS founder says Cursor AI agent deleted its production database in 9 seconds after misusing a root API token, exposing major Railway security flaws.
Microsoft’s April update to Visual Studio introduces cloud agent integration in GitHub Copilot, enabling developers to offload tasks to remote infrastructure for scalable, isolated execution. You can now start cloud agent sessions directly from Visual Studio. Custom agents now support user-level definitions that persist across projects, making it easier to reuse configurations. The update also…
COMPANY NEWS: AI agents in GitLab Duo Agent Platform can now call Anthropic’s newest Claude models, with every action governed by GitLab’s existing compliance, audit, and policy framework; no separate governance layer required. Enterprises can access Claude models, including the newly released Claude Opus 4.7, through GitLab via Google Cloud and AWS Bedrock to route…
Cequence Security has announced the general availability of Agent Personas in Cequence AI Gateway. These capabilities give enterprises granular, infrastructure-level control over what AI agents can do, down to individual tool calls, closing a critical privilege gap that identity alone cannot address. As organizations deploy AI agents to connect to enterprise applications via the Model…
Microsoft fixed a Microsoft Entra ID flaw where the Agent ID Administrator role could enable privilege escalation and account takeover. Microsoft addressed a flaw in Microsoft Entra ID that could let attackers take over service accounts. The issue involved the Agent ID Administrator role, which manages AI agent identities and access, and could be abused…
Microsoft Entra Agent ID flaw allowed privilege escalation and tenant takeover via Service Principal abuse, now fully patched by Microsoft.
Microsoft has beefed up Copilot’s capabilities in Word, Excel and PowerPoint, claiming its Agent Mode will help speed up workers’ output. The new features, announced last year, mean that Copilot can work more efficiently with Office applications, for example, understanding the richness of a pivot table in Excel or the use of animations in PowerPoint. In…
The AI Agent Authority Gap – From Ungoverned to Delegation As discussed in our previous article, AI agents are exposing a structural gap in enterprise security, but the problem is often framed too narrowly. The issue is not simply that agents are new actors. It is that agents are delegated actors. They do not emerge…
The AI agent introduction frenzy continued at a torrid pace this week, with OpenAI launching what it called workspace agents in ChatGPT and Microsoft adding hosted agents to its Foundry Agent Service. Both launched on the same day that Google both updated its Gemini Enterprise app to provide new ways for office workers to build,…
Google Cloud will attribute a unique cryptographic ID every AI agent that will be tied to “traceable and auditable” authorization policies
New Rubrik Agent Cloud Integration with Gemini Enterprise Agent Platform provides a unified control layer for AI agent operations
Tencent has opened an international beta of QClaw, an AI agent application aimed at consumers in Canada, Japan, Singapore, South Korea, and the United States. The first wave is capped at 20,000 users. Additional markets are scheduled to follow. QClaw runs on Windows and MacOS. Setup takes about three minutes and involves downloading the application,…
COMPANY NEWS: Customers can route GitLab Duo Agent Platform inference through Amazon Bedrock models already running in their AWS accounts without new vendor onboarding or model endpoints. GitLab Credits purchased through AWS Marketplace count toward existing AWS spending commitments. GitLab’s Bring Your Own Model (BYOM) capability for Self-Managed customers lets teams connect their self-hosted AI…
A high-severity authentication flaw in Microsoft’s Azure SRE Agent exposed sensitive agent data to unauthorized network access, according to a confirmed vulnerability disclosure. The issue was identified by Enclave AI researcher Yanir Tsarimi, who detailed the findings in a blog post describing how agent interactions could be accessed without proper authentication controls. The vulnerability has…
Agent delivers seamless integration in minutes using natural language – enabling high-stakes organisations of all sizes to move faster and tech teams to prove concepts quickly
AI agent building tools enable users to configure Model Context Protocol (MCP) servers may be exposing systems to remote code execution due to an architectural decision in Anthropic’s reference implementation. At issue are unsafe defaults in how MCP configuration works over the STDIO interface, with broad implications for the agent ecosystem, according to a new…
The EU AI Act is 144 pages long. The logging requirements that matter for AI agent developers sit across four articles that keep referencing each other. Here’s what they say, when the deadlines hit, and where the gaps are. Your agent is probably high-risk The Act doesn’t mention “AI agents” by name. What matters is…
Broadcom has announced VMware Tanzu Platform agent foundations, introducing a secure-by-default agentic runtime designed to accelerate the delivery of autonomous AI applications. By extending the trusted code-to-production simplicity of Tanzu Platform to AI agents, Broadcom is enabling enterprise developers to move beyond siloed AI experiments and into scalable, governed production on VMware Cloud Foundation (VCF).…
COMPANY NEWS: AI agents in GitLab Duo Agent Platform can now call foundation models through Vertex AI, including Gemini models, with agent actions governed by GitLab’s built-in compliance and audit controls. Organisations can run GitLab’s AI Gateway on Google Cloud with no separate AI infrastructure to provision or manage. Customers with Google Cloud commitments can…
Cloudflare Mesh helps any developer encrypt every human, code, and agent connection point without ever exposing internal infrastructure and data to the public Internet
Microsoft has quietly introduced the Agent Governance Toolkit, an open-source project designed to monitor and control AI agents during execution as enterprises try to move them into production workflows. The toolkit, which is a response to the Open Worldwide Application Security Project’s (OWASP) emerging focus on AI and LLM security risks, adds a runtime security…
Microsoft has quietly introduced the Agent Governance Toolkit, an open-source project designed to monitor and control AI agents during execution as enterprises try to move them into production workflows. The toolkit, which is a response to the Open Worldwide Application Security Project’s (OWASP) emerging focus on AI and LLM security risks, adds a runtime security…
The recent incident at Meta, where an AI agent exposed sensitive internal data following a routine query, is a timely reminder that AI risk is already operational.
Cloud storage company Box is announcing the launch of Box Agent, an AI-powered agent with reasoning and task-completion capabilities, along with enhancements to Box AI Studio. Box Agent promises context-based capabilities to iterate on corporate workflows The Box Agent acts as a unified AI engine across Box and autonomously understands a user’s intent from their…
Exabeam has announced the expansion of Exabeam Agent Behavior Analytics (ABA). Without direct visibility into how employees use AI assistants, what they query, what data they share, how frequently they interact, and from where, organizations cannot establish a baseline for normal AI behavior, investigate potential misuse, or detect emerging agentic insider threats. New support to…
Exabeam has recently announced the expansion of Exabeam Agent Behavior Analytics (ABA), extending behavior detection and response to OpenAI ChatGPT and Microsoft Copilot. Capabilities close the visibility gap in AI usage The expansion applies behavior profiling and analytics to the digital workforce, as organizations don’t have direct visibility into how employees are using AI assistants,…
AWS Security Agent on-demand penetration testing is now generally available, enabling you to run comprehensive security tests across all your applications, not only your most critical ones. This milestone transforms penetration testing from a periodic bottleneck into an on-demand capability that scales with your development velocity across AWS, Azure, GCP, other cloud-providers, and on-premises. With…
Amazon’s latest AI capabilities bring on-demand penetration testing through the AWS Security Agent, alongside the AWS DevOps Agent. “These agents are changing the way we secure and operate software. AWS Security Agent compresses penetration testing timelines from 2-6 weeks to 1-2 days. AWS DevOps agent gives teams 3–5x faster incident resolution so they can spend…
AI agent risk isn’t equal, it scales with access to systems and level of autonomy. Token Security explains how CISOs should categorize agents and prioritize what to secure first. […]
Microsoft is expanding its Microsoft 365 Copilot “Researcher” agent with new multi-model capabilities designed to improve the accuracy and depth of AI-generated research outputs. The update introduces a “Critique” system that assigns separate roles for generation and evaluation, alongside a “Council” feature that compares outputs from multiple models and highlights agreement, divergence, and unique insights.…
Container sandboxes are part of routine AI agent testing and deployment. Agents use them to run code, edit files, and interact with system resources without direct access to the host. The SandboxEscapeBench benchmark, developed by researchers at the University of Oxford and the AI Security Institute, evaluates whether an agent with shell access can escape…
Vorlon has unveiled AI Agent Flight Recorder and AI Agent Action Center, adding forensics and coordinated response to secure enterprise agentic ecosystems and close a key security gap. The agentic ecosystem contains SaaS applications, AI agents, API integrations, non-human identities, and the sensitive data flows connecting them. It’s become the fastest-growing attack surface in the…
Domain-Specific Small Language Model Accelerates Trusted AI Agent Deployment and Control
Cisco has introduced solutions to address AI security issues and remove a top barrier to agent adoption. By establishing trusted identities, enforcing strict zero trust Access controls, hardening agents before deployment, enforcing guardrails at runtime, and giving SOC teams the tools to stop threats at machine speed, Cisco is building security into the foundation of…
As CISOs worry about AI agent sprawl, Palo Alto Networks has announced an update to its Prisma AIRS security platform and enterprise browser to include the ability to discover AI agents, models, and connections across the entire IT environment, to scan agents for vulnerabilities, and to allow admins to simulate red team tests for agents.…
COMPANY NEWS: New capabilities in the BeyondTrust Pathfinder Platform secure AI agent coworkers on endpoints and AI agent workloads across cloud infrastructure and SaaS platforms Security teams gain visibility into AI agent identities, privileges, and secrets across platforms including OpenAI, AWS Bedrock, Salesforce Agentforce, ServiceNow, and Google Vertex AI
New AI agent automates investigations with senior SOC analyst expertise at machine scale and speed to deliver accurate, fully explained verdicts that dramatically reduce remediation times
Apiiro has announced AI Threat Modeling, a new capability within Apiiro Guardian Agent that automatically generates architecture-aware threat models to identify security and compliance risks before code exists. AI Threat Modeling allows enterprises to prevent risks at the speed of AI, whether developing first-party applications, delivering third-party applications to the cloud, or addressing the new…
Astrix Security has revealed a major expansion of its AI agent security platform, covering every layer where AI agents operate in the enterprise: from managed AI platforms to shadow deployments running on managed devices, detecting both agent existence and unauthorized access to enterprise resources, and enforcing policy over what agents are allowed to do. AI…
Agent Val exploits, mitigates and revalidates exposures against compensating controls continuously, dramatically reducing the average window of exposure
Discover why 85% of organizations are exploring agentic AI, yet only 5% are in production. Learn how to bridge the agent trust gap with robust security.
COMPANY NEWS: Organisations on the GitLab.com free tier can now start using GitLab Duo Agent Platform by purchasing a monthly commitment of GitLab Credits, giving every team access to agentic AI across the full software lifecycle. Agentic code reviews now cost a flat $0.25 per review (4 code reviews per GitLab Credit today), making automated…
Token Security has unveiled intent-based AI agent security, a new approach that governs autonomous agents in enterprise environments by aligning their permissions with their intended purpose. As organizations deploy autonomous AI agents across enterprise infrastructure, security models are struggling to contain the risks. Token Security has been advancing the concept of intent-based security for AI…
Dropzone AI has released the AI Threat Hunter, its newest AI agent that enables security teams to proactively search for threats across their environments around the clock. The AI Threat Hunter is the next agent joining the Dropzone’s Agentic SOC team, expanding what AI agents can do across the full spectrum of detection and response.…
SaaS security platform Reco has decided to address the “agent sprawl” challenge from the increased adoption of AI-driven tools by enterprises. It argues that enterprises are faced with a security situation as numerous autonomous agents now traverse multiple systems, accessing sensitive data, and executing actions without direct human oversight. To help contain this risk, the…
Virtue AI has announced Agent ForgingGround with built-in Red-Teaming Agents, the first enterprise-scale testing ground designed to continuously evaluate and stress-test AI agents (including multi-agent systems) before, during, and after deployment. As organizations adopt large-scale AI agents, many enterprises are unprepared to manage agent risk. AI agents can call tools, access sensitive data, and execute…
Secure Code Warrior has announced SCW Trust Agent: AI, a governance solution designed to make AI influence in software development visible, attributable, and enforceable at the point of commit, enabling enterprises to scale AI coding tools with measurable control over software risk. Organizations can trace which AI models influenced specific commits, correlate that influence with…
Kore.ai has announced the launch of its Agent Management Platform, a unified command center designed to govern, monitor, and manage AI agents and AI systems across the enterprise. As organizations deploy AI and multi-agent systems, many are encountering “AI sprawl,” a phenomena analysts describe as dozens of AI initiatives across different teams, tools, and clouds…
Jozu has announced the launch of Jozu Agent Guard, a zero-trust AI runtime that executes agents, models, and MCP servers in secure environments with built-in policy enforcement and guardrails that cannot be disabled. As enterprises adopt AI agents, MCP servers, and tools such as Copilot, OpenClawd, and Claude Code, security teams face a growing gap.…
What happens when an autonomous AI agent is turned loose on another autonomous AI agent? It chains together bugs that humans would consider benign, easily bypasses authentication controls, and even unexpectedly masquerades as Donald Trump to get its way. This was what CodeWall found in a recent red-teaming experiment when it pitted its autonomous AI…
A recent research paper describing the training of an experimental AI agent has started a discussion after the…
Singulr AI has announced the launch of Agent Pulse, extending its Unified AI Control Plane to autonomous AI agents and model context protocol (MCP) servers. Agent Pulse delivers enforceable runtime governance, contextual discovery, and measurable oversight for the agentic enterprise. Agent Pulse serves as an extension of the Unified AI Control Plane for the agentic…
Learn about OpenClaw (a sovereign agent) and how this can be viable for enterprises.
OpenAI’s new AppSec agent, Codex Security, has already flagged over 11,000 high-severity and critical flaws in real-world codebases during its first 30 days of research testing. The tool, designed to automatically find, validate, and fix vulnerabilities in software repositories, reportedly identified about 800 critical issues in more than a million scanned commits. According to an…
Security firm Pillar reveals the Chaos Agent in which Hackerbot-Claw, an AI agent, used natural language to compromise major GitHub projects and hijack developer tools.
OpenAI introduced Codex Security, an AI agent that reviews codebases to find, verify, and help fix software vulnerabilities. The launch comes a few weeks after rival Anthropic unveiled its Claude Code Security tool. The feature is available in research preview via Codex Web for ChatGPT Pro, Enterprise, Business, and Edu customers, with free access for…
Cursor Automations, the always-on agent platform from Cursor, is expanding with a new generation of autonomous systems that streamline code review, incident response, and other engineering workflows. The platform runs AI agents on schedules or in response to development events. These triggers include merged pull requests on GitHub, newly created issues in Linear, messages sent…
DeepKeep on Tuesday introduced a new AI Agent Scanner designed to help enterprises identify and secure the growing attack surface created by AI agents embedded in business workflows. The Tel Aviv-based AI security vendor said its latest release provides structured attack surface scanning and discovery for agentic AI environments, where large language model (LLM)-based agents…
Security researchers have disclosed a high-severity vulnerability dubbed “ClawJacked” in the popular AI agent OpenClaw that allowed a malicious website to silently bruteforce access to a locally running instance and take control over it. […]
If you thought running an AI agent locally kept it safely inside your machine’s walls, you’re in for a surprise. Researchers at Oasis Security have disclosed a flaw chain that allowed a malicious website to quietly connect to a locally running OpenClaw agent and take full control. The issue stems from a fundamental assumption baked…
COMPANY NEWS: New virtual agent capabilities reduce customer effort, prevent repeat contacts, and give service leaders confidence to scale automation With 43% of consumers saying chatbots fail to resolve their issues, Zoom Virtual Agent 3.0 can help organisations close that gap
Veeam Software announced Agent Commander, a unified solution to help organizations safely detect AI risk, protect AI systems, and undo AI mistakes, enabling them to proactively address AI-driven risks and securely scale AI agents everywhere. The first integration from Veeam’s acquisition of Securiti AI, Agent Commander combines the capabilities of both to give organizations visibility,…
New AI Agent From KnowBe4 Cuts Security Training Administration From Hours to Seconds
A new security bypass has users installing AI agent OpenClaw — whether they intended to or not. Researchers have discovered that a compromised npm publish token pushed an update for the widely-used Cline command line interface (CLI) containing a malicious postinstall script. That script installs the wildly popular, but increasingly condemned, agentic application OpenClaw on…