All news with #agentic ai tag

🔎 A Sygnia report details how a lone threat actor leveraged AI to complete in 72 hours what would normally take weeks, using established cloud attack techniques rather than novel exploits. The attacker obtained an AWS access key via an internet-facing app and used agentic AI workflows to search for secrets, establish persistence, exfiltrate RDS data, and perform impact actions. The report highlights gaps in secrets management, identity governance, deployment workflows and visibility, and provides containment recommendations for defenders.

🛡️ The Five Eyes recently warned that AI models increasingly enable autonomous cyberattacks, amplifying risks long present in cyberspace. Bruce Schneier argues that AI widens the gap between skill and ability: tools let less-skilled actors cause damage once limited to experts. He warns guardrails from large vendors won’t stop open-source or locally run models and urges using AI defensively to detect, remediate, and respond faster to evolving threats.

🔒 The UK National Cyber Security Centre (NCSC) has launched the Cyber Shield initiative to build a national cyber-defence capability powered by agentic AI. The project will use coordinated red and blue agents to discover and mitigate vulnerabilities at scale, enable national automated scanning, and support real-time intelligence sharing. The NCSC says success requires partnerships with government, critical infrastructure and frontier AI providers.

🧭 This guide explains how to build, register, and commercialize AI agents as Agents-as-a-Service (AaaS) for deployment in the Gemini Enterprise app and Google Cloud Marketplace. It details required artifacts like the A2A Agent Card, authentication options including OAuth and Dynamic Client Registration (DCR), and the integration points for procurement, billing, and entitlement management. The article also outlines the seller journey, testing, and the end-to-end procurement and registration lifecycle across Billing Administrator, Discovery Engine Administrator, and end-user roles.

🧭 This Google Cloud blog summarizes findings from a survey of over 1,400 IT leaders showing a widening gap between AI ambitions and existing infrastructure. It explains why agentic AI stresses legacy systems and highlights needs for fluid compute, unified data layers, hybrid deployments, and energy-efficient co-designed silicon like TPU 8t. The post also promotes Google Cloud’s AI Hypercomputer and governance solutions such as Agent Gateway.

🤖 This post introduces the Gemini Enterprise Agent Platform and offers 20 practical questions for IT and engineering leaders to consider when building AI agents. It covers who builds agents, which development tools to use, how to connect data and other agents, strategies for scaling, and methods for securing execution and preserving context. The article pairs guidance with recommended Google tools like ADK, Antigravity, Agent Runtime, MCP, A2A, and Agent Studio.

🛡️ Zscaler tested 26 LLMs and found several autonomous agents susceptible to indirect prompt injection (IPI) traps, with some high-end models failing while a few lower-tier models fared better. The vendor identified hidden instructions on websites that manipulated agent behavior and caused real-world impacts in controlled tests. Experts warn that agent risk is dynamic, the attack surface is architectural, and binary "safe/vulnerable" labels are overly simplistic for CISOs. The findings highlight that agentic AI introduces new trust boundaries and insider-like threats to enterprise security.

🛡️ Zscaler tested 26 LLM-based autonomous agents and found several susceptible to indirect prompt injection (IPI) schemes, with some high-end models failing while a few lower-tier models fared better. The vendor reported four models as "vulnerable" and three as "safe," but experts warn that agent behavior evolves and binary classifications can be misleading. The findings highlight the architectural risks in agentic AI where untrusted content in the context window can be treated as authoritative, expanding the attack surface for enterprises.

🔒 This post describes a reference implementation using Cedar on AWS to prevent silent privilege escalation in multi-agent AI delegation chains. It outlines a three-layer policy model—agent-to-tool, agent-to-agent delegation, and originating user authorization—using verified token claims and HMAC-signed context. The architecture uses an MCP adapter Lambda and a Cedar evaluator Lambda to enforce policies sequentially and halt on the first deny. It includes schema, entity registrations, policy examples, deployment steps, and end-to-end test scenarios demonstrating how the model enforces least privilege.

🔒 Google Cloud and Valtech introduce Nexus SDV, an open-source, modular platform that enables AI-native, scalable management of software-defined vehicles. The platform integrates with Android Automotive OS and supports up to 100 million devices while emphasizing TCO reduction via Arm-based compute and Bigtable. Nexus AI leverages Gemini models and the Gemini Enterprise Agent Platform for real-time telemetry analysis and agentic vehicle capabilities. Security is built-in with mTLS/PKI, identity brokering, secret management, network isolation, and an enterprise Secure AI Framework.

🔍 Zscaler ThreatLabz uncovered real-world campaigns using indirect prompt injection, where hidden instructions embedded in web pages steer AI agents. Attackers used SEO poisoning to surface malicious pages and hid prompts via CSS and JSON-LD metadata. One campaign impersonated a Python library to trick agents into paying a $3 bogus API key; another typosquatted a DeBank site to claim authority. Tests across 26 LLMs showed varying susceptibility depending on model and context.

🤖 Stephen Wilson of HashiCorp describes agentic AI as “really smart kindergartners” — capable of execution but lacking judgment. This mismatch strains traditional zero trust models that authenticate humans and grant privileges gradually, because agents can be created and destroyed rapidly. Organizations often respond by lowering controls, risking incidents such as accidental deletion of production data. Wilson argues this will force necessary long-term improvements like zero standing privilege and dynamic credentials while keeping humans "on the loop."

🛡️ Existing security controls weren’t built for autonomous AI agents, and static credentials and standing privileges are insufficient. Organizations must define agentic identity, secure agent-to-agent communication, adopt dynamic secrets management, enforce least privilege for delegated workflows, and unify workforce identity. Governance across the identity lifecycle is essential to ensure auditable, revocable, and context-aware access for agents.

🤖 Stephen Wilson of HashiCorp explains how enterprise AI is evolving from human-assisted tools to autonomous agents and operators, and why governance must mature accordingly. He describes three adoption patterns—AI as assistant, AI as agent, and AI as operator—and details the increasing needs for identity, access controls, auditability, and accuracy at each stage. As organizations grant agents more autonomy, security controls must expand from user-level boundaries to team and organizational governance.

🔒 Sysdig reports a novel ransomware campaign, dubbed JadePuffer, driven entirely by a large language model agent that exploited CVE-2025-3248 in an internet-facing Langflow instance. The automated attack conducted reconnaissance, credential harvesting, lateral movement, and destructive actions against production databases, encrypting and deleting Nacos configurations so they could not be recovered. Sysdig highlights automation of old vulnerabilities, agent narration that may aid detection, and the erosion of response time for defenders.

🔒 This article explains how traditional identity lifecycle management — built around HR-driven joiner, mover, and leaver events — fails to govern AI agents. It describes how agents are created outside HR and IGA workflows, arrive with embedded credentials, and expand access dynamically at runtime. The piece highlights gaps in provisioning, access reviews, and offboarding when agents proliferate across parallel instances and orchestration layers.

🛡️ Researchers discovered two vulnerabilities in the Cursor AI-enabled IDE that enable prompt-injection-driven remote code execution by escaping the command execution sandbox. The flaws, CVE-2026-50548 and CVE-2026-50549, allow attackers to change the working directory and exploit symlink canonicalization fallbacks to write or overwrite files outside the project scope. Cursor patched the issues in version 3.0, and the findings underscore broader risks in agentic AI workflows and the difficulty of defending against prompt injection.

🛠️ This post outlines an architecture using Google's ADK and Gemini Flash to replace static prompt-driven agents with a just-in-time, metadata-driven orchestration. It externalizes JSON schema descriptors to a Central Metadata Registry and employs a lightweight discovery prompt plus dynamic validation hooks (Cloud Run) to ensure deterministic, schema-compliant payloads. The pattern reduces context bloat, lowers token costs, and prevents attention diffusion in multi-agent workflows.

🛡️ Cloudflare updates its bot management to distinguish between three AI use cases—Search, Agent, and Training—so site owners can better control access and compensation for their content. The company will change defaults on September 15, 2026, blocking Training and Agent bots on ad-bearing pages while leaving Search allowed. Cloudflare also launched BotBase, a searchable directory of tracked bots, and added a new content-use signal for robots.txt to express preferences like use=reference.

🧭 Cloudflare reports a rapid shift toward an agent-driven Internet where AI training and mixed-use crawlers dominate. Publishers face falling referral traffic as over 50% of Internet traffic is now non-human, and AI companies increasingly ingest content without compensation. Cloudflare highlights tools and marketplace developments that restore publisher control, enable attribution, and support licensing between content owners and AI firms.