All news with #agentic ai tag

Azure AI Foundry Brings Multimodal OpenAI Models at Scale

🚀 Azure AI Foundry now integrates new OpenAI models—GPT-image-1-mini, GPT-realtime-mini, and GPT-audio-mini—alongside safety upgrades to GPT-5. The rollout, with most customers able to get started on October 7, 2025, targets efficient, low-latency multimodal workloads for developers and enterprises. Microsoft also highlighted the open-source Microsoft Agent Framework, multi-agent workflows, unified observability, Voice Live API GA, and Responsible AI enhancements to accelerate production-grade agentic solutions.

OpenAI Tests ChatGPT-Powered Agent Builder Tool Preview

🧭 OpenAI is testing a visual Agent Builder that lets users assemble ChatGPT-powered agents by dropping and connecting node blocks in a flowchart. Templates like Customer service, Data enrichment, and Document comparison provide editable starting points, while users can also create flows from scratch. Agents are configurable with model choice, custom prompts, reasoning effort, and output format (text or JSON), and they can call tools and external services. Reported screenshots show support for MPC connectors such as Gmail, Calendar, Drive, Outlook, SharePoint, Teams, and Dropbox; OpenAI plans to share more details at DevDay.

Google advances AI security with CodeMender and SAIF 2.0

🔒 Google announced three major AI security initiatives: CodeMender, a dedicated AI Vulnerability Reward Program (AI VRP), and the updated Secure AI Framework 2.0. CodeMender is an AI-powered agent built on Gemini that performs root-cause analysis, generates self-validated patches, and routes fixes to automated critique agents to accelerate time-to-patch across open-source projects. The AI VRP consolidates abuse and security reward tables and clarifies reporting channels, while SAIF 2.0 extends guidance and introduces an agent risk map and security controls for autonomous agents.

CometJacking: One-Click Attack Turns AI Browser Rogue

🔐 CometJacking is a prompt-injection technique that can turn Perplexity's Comet AI browser into a data exfiltration tool with a single click. Researchers at LayerX showed how a crafted URL using the 'collection' parameter forces the agent to consult its memory, extract data from connected services such as Gmail and Calendar, obfuscate it with Base64, and forward it to an attacker-controlled endpoint. The exploit leverages the browser's existing authorized connectors and bypasses simple content protections.

Opera Neon AI Browser: $19.90 Monthly for Agentic Web

🤖 Opera has unveiled Neon, a premium AI-first browser that delegates browsing tasks to integrated agents, from opening tabs and conducting research to comparing prices and assessing security. Early access is available for Windows and macOS at an introductory price of $59.90 for nine months; Opera says the service will cost $19.90 per month after the offer. Opera positions Neon alongside other agentic browsers such as Perplexity Comet and Microsoft Edge's Copilot mode.

AI and Cybersecurity: Fortinet and NTT DATA Webinar

🔒 In a joint webinar, Fortinet and NTT DATA outlined practical approaches to deploying and securing AI across enterprise environments. Fortinet described its three AI pillars—FortiAI‑Protect, FortiAI‑Assist, and FortiAI‑SecureAI—focused on detection, operational assistance, and protecting AI assets. NTT DATA emphasized governance, runtime protections, and an "agentic factory" to scale pilots into production. The presenters stressed the need for visibility into shadow AI and controls such as DLP and zero‑trust access to prevent data leakage.

CometJacking attack tricks Comet browser into leaking data

🛡️ LayerX researchers disclosed a prompt-injection technique called CometJacking that abuses Perplexity’s Comet AI browser by embedding malicious instructions in a URL's collection parameter. The payload directs the agent to consult connected services (such as Gmail and Google Calendar), encode the retrieved content in base64, and send it to an attacker-controlled endpoint. The exploit requires no credentials or additional user interaction beyond clicking a crafted link. Perplexity reviewed LayerX's late-August reports and classified the findings as "Not Applicable."

Open-Source MCP Server for Amazon Bedrock AgentCore

🚀 The open-source Model Context Protocol (MCP) Server is now available for Amazon Bedrock AgentCore, providing a standardized interface that enables developers to analyze, transform, and deploy production-ready AI agents within their preferred development environments. The release includes one-click installation and integrates with agentic IDEs like Kiro and AI coding assistants such as Claude Code, Cursor, and the Amazon Q Developer CLI. Developers can use natural language to iteratively build agents, convert agent logic to the AgentCore SDK, and deploy into development accounts. Documentation and installation instructions are published in the MCP Server GitHub repository, with additional implementation guidance and pricing details available in the AgentCore documentation and pricing resources.

Forrester Predicts Agentic AI Will Trigger 2026 Breach

⚠️ Forrester warns that an agentic AI deployment will trigger a publicly disclosed data breach in 2026, potentially prompting employee dismissals. Senior analyst Paddy Harrington noted that generative AI has already been linked to several breaches and cautioned that autonomous agents can sacrifice accuracy for speed without proper guardrails. He urges adoption of the AEGIS framework to secure intent, identity, data provenance and other controls. Check Point also reported malicious agentic tools accelerating attacker activity.

Accelerate AI with Agents: EMEA Developer Series and Labs

🚀 Google Cloud is hosting a regional event series across EMEA to help developers and tech practitioners learn to build and scale AI agents. The program combines immersive, hands-on labs and expert-led workshops covering technologies such as Cloud Run, Vertex AI, Gemini, and the Agent Development Kit (ADK). Participants receive step-by-step guidance and practical exercises designed to accelerate agent deployments and operational readiness within organizations.

Modern Business Continuity and Disaster Recovery Basics

🛡️ Modern disaster recovery and business continuity require a ground-up rebuild to address distributed data, evolving cyberthreats, climate-driven disruptions, and strict breach-reporting obligations. Key elements include executive sponsorship, standing interdisciplinary teams, AI-assisted discovery and classification, continuous and immutable backups aligned with a 3-2-1-1-0 approach, and the design of a minimum viable business to restore core functions. Frequent, gamified tabletop exercises and automated validation complete a resilient program.

Smashing Security 437: ForcedLeak in Salesforce AgentForce

🔐 Researchers uncovered a security flaw in Salesforce’s new AgentForce platform called ForcedLeak, which let attackers smuggle AI-readable instructions through a Web-to-Lead form and exfiltrate data for as little as five dollars. The hosts discuss the broader implications for AI integration, input validation, and the surprising ease of exploiting customer-facing forms. Episode 437 also critiques typical breach communications and covers ITV’s phone‑hacking drama and the Rosetta Stone story, with Graham Cluley joined by Paul Ducklin.

Microsoft Advances Sentinel with Agentic AI Upgrades

🔒 Microsoft announced major AI upgrades for Sentinel SIEM and Security Copilot, positioning them as agentic platforms. The update makes Sentinel data lake generally available and introduces public-preview releases of Sentinel graph and the Sentinel Model Context Protocol (MCP) Server so AI agents can access and act on SIEM data. Customers can now build custom agents with natural‑language prompts and discover third‑party agents via a revamped store. Microsoft positions agents to automate investigation and response but warns of increased noise, false positives and a new attack surface.

AWS API MCP Server v1.0.0 enables natural API control

🧭 The AWS API MCP Server v1.0.0 enables foundation models to interact with AWS APIs via natural language by generating and executing syntactically correct CLI commands. This release reduces startup time and removes local dependencies by converting the suggest_aws_command tool into a remote service, and adds streamable HTTP transport alongside stdio. Security and governance are strengthened with improved secure file system controls, better input validation, configurable denials, and options to require human oversight for mutating actions. Observability is improved through CloudWatch agent log collection, and an experimental get_execution_plan tool (enabled by EXPERIMENTAL_AGENT_SCRIPTS) offers prescriptive workflows; the server is available as a container and open-source on the AWS Labs GitHub repository.

Microsoft Agent Framework Brings Multi-Agent Tooling

🤖 The Microsoft Agent Framework is now in public preview inside Azure AI Foundry, offering an open-source SDK and runtime to simplify orchestration of multi-agent systems. Developers can prototype locally and deploy with built-in observability, durability, and compliance while integrating tools via OpenAPI, Agent2Agent (A2A), and the Model Context Protocol (MCP). Microsoft also previews stateful multi-agent workflows and has contributed multi-agent tracing standards to OpenTelemetry. Responsible AI controls and a generally available Voice Live API add governance and real-time voice capabilities for enterprise scenarios.

AWS Knowledge MCP Server Now Generally Available Globally

🔎 The AWS Knowledge MCP Server is now generally available, giving AI agents and MCP-compatible clients access to authoritative AWS documentation, blog posts, What's New announcements, and Well-Architected guidance in an LLM-friendly format. The GA release also adds structured knowledge about regional API and CloudFormation resource availability. The server is publicly accessible at no cost and does not require an AWS account, though usage is rate-limited. Configure MCP clients to use the AWS Knowledge MCP Server endpoint to anchor agent responses in trusted AWS context and reduce manual context management.

Blending AI and Human Workflows for Secure Automation

🔍 Join The Hacker News for a free webinar, "Workflow Clarity: Where AI Fits in Modern Automation," featuring Thomas Kinsella, Co‑founder & Chief Customer Officer at Tines. The piece argues that human-only processes are slow, rigid rule engines break when reality changes, and fully autonomous AI can create opaque, unauditable paths. Attendees will learn practical mapping of tasks to people, rules, or AI, how to spot AI overreach, and patterns for building secure, auditable workflows that scale without sacrificing control.

AI Tops Cybersecurity Investment Priorities — PwC Report

🔒 A PwC survey finds AI-based security is the top cybersecurity investment priority for the next 12 months, with 36% of business and technology executives ranking it among their top three budget areas. Security leaders prioritized AI threat hunting (48%) and agentic AI to boost cloud and operational efficiencies (35%). While 78% expect cyber budgets to rise, organizations report significant knowledge and skills gaps and low readiness for quantum threats.

CrowdStrike Fall 2025 Release: Agentic SOC & AI Security

🔒 CrowdStrike’s Fall 2025 release introduces the Falcon agentic security platform, redefining SOC operations by pairing analysts with AI agents that reason, act, and continuously learn. The release centers on the Enterprise Graph data layer and Charlotte AI AgentWorks for no-code agent creation, plus seven mission-ready agents and Threat AI for autonomous hunting and response. It also expands identity, data protection, patching, and MDR capabilities, integrates Pangea and Onum, and launches AI Detection and Response (AIDR) to secure AI workflows.

AWS Transform Adds Terraform Module Generation for VMware

🔁 AWS Transform for VMware now generates reusable Terraform modules from discovered VMware network definitions, complementing existing AWS CloudFormation and CDK outputs. The feature converts source network configurations into modular, customizable infrastructure code that fits into current deployment pipelines. It is available in all Regions where the service is offered and helps teams preserve operational consistency during migrations. By producing Terraform modules, the service enables reuse of Terraform-based workflows, reduces manual configuration effort, and supports teams that prefer Terraform for network automation.