OpenAI Agents SDK Improves Governance with Sandbox Execution
OpenAI has launched sandbox execution within its Agents SDK, giving enterprises a model‑native harness that isolates code, credentials, and data. The new infrastructure lets developers run autonomous agents in controlled containers, connect to major cloud storage, and snapshot state for rapid recovery. Oscar Health used the SDK to automate a complex clinical‑records workflow, extracting metadata and encounter boundaries more reliably. The capabilities are GA for Python developers, with TypeScript support planned later.
Commvault Launches a ‘Ctrl-Z’ for Cloud AI Workloads
Commvault unveiled AI Protect, a cloud‑native solution that detects, monitors, and reverses actions taken by autonomous AI agents in AWS, Azure and Google Cloud. The platform continuously scans for hidden or "shadow" AI workloads, logs every API call, and can...
Citizen Developers Now Have Their Own Wingman
Emergent has launched Wingman, an autonomous AI agent platform that lets citizen developers create and manage software applications without coding. The service can control popular tools such as WhatsApp, Telegram, email and GitHub, and separates tasks into “trust boundaries” that...
Drones Get Smarter for Large Farm Holdings
Singapore’s DroneDash Technologies and GEODNET have created a joint venture, GEODASH Aerosystems, to launch an AI‑driven agricultural spraying drone. The platform combines DroneDash’s computer‑vision system with GEODNET’s centimetre‑level positioning, eliminating the need for pre‑flight field mapping and enabling real‑time altitude...
SAP Brings Agentic AI to Human Capital Management
SAP announced its SuccessFactors 1H 2026 release, embedding agentic AI across recruiting, payroll, workforce administration and talent development. The AI agents continuously monitor system states, detect data anomalies and suggest context‑aware fixes, cutting mean‑time‑to‑resolution for internal support tickets. Integrated semantic search...
Canada’s Scotiabank Preps for Its AI Future
Scotiabank has unveiled Scotia Intelligence, an AI framework that consolidates data platforms, governance, and software tools into a single environment. The employee‑focused component, Scotia Navigator, lets staff create and deploy AI assistants while staying within strict security and regulatory rules....
Hyundai Expands Into Robotics and Physical AI Systems
Hyundai Motor Group announced a $26 billion U.S. investment through 2028 to develop physical AI and robotics, building on $20.5 billion invested over the past four decades. The company aims to produce up to 30,000 collaborative robots annually by 2030, focusing on...
Strengthening Enterprise Governance for Rising Edge AI Workloads
Google’s release of Gemma 4, an open‑weight model designed for on‑device execution, is upending traditional enterprise AI security. The model lets engineers run sophisticated autonomous agents directly on laptops and edge hardware, sidestepping cloud firewalls and API gateways. This creates a...
Asylon and Thrive Logic Bring Physical AI to Enterprise Perimeter Security
Asylon and Thrive Logic have partnered to embed physical AI into enterprise perimeter security, combining Asylon’s autonomous robotic patrols with Thrive Logic’s AI‑driven analytics and automated incident workflows. The integration streams video from mobile robots to the AI platform, which...
Boomi Calls It “Data Activation” And Says It’s the Missing Step in Every AI Deployment
Boomi warns that fragmented, poorly‑labelled data is the biggest obstacle to enterprise AI in 2026. The company tracks 75,000 AI agents in production across more than 30,000 customers, including over a quarter of the Fortune 500. Its March 9 platform update...
Anthropic’s Refusal to Arm AI Is Exactly Why the UK Wants It
Anthropic refused the U.S. Pentagon’s demand to strip ethical guardrails from its Claude model, prompting Washington to blacklist the firm and cancel a $200 million contract. In response, the United Kingdom’s Department for Science, Innovation and Technology has drafted a package—including...
KiloClaw Targets Shadow AI with Autonomous Agent Governance
KiloClaw has launched an enterprise‑grade platform that gives security teams visibility and control over autonomous agents operating outside official IT channels. The tool addresses the growing “Bring Your Own AI” (BYOAI) trend, where developers and knowledge workers deploy personal scripts...
5 Best Practices to Secure AI Systems
As AI becomes integral to critical operations, its unique attack surface outpaces traditional security measures. Experts recommend a multi‑layered strategy that starts with strict access controls and encryption, followed by AI‑specific defenses such as firewalls and red‑team testing. Continuous visibility...
Experian Uncovers Fraud Paradox in Financial Services’ AI Adoption
Experian’s 2026 Future of Fraud Forecast warns that AI agents, once a defensive tool for banks, are being weaponised by fraudsters, creating a machine‑to‑machine mayhem where liability is unclear. FTC data shows consumers lost over $12.5 billion to fraud in 2024,...
Autonomous AI Systems Depend on Data Governance
The focus of AI safety is shifting from model‑centric controls to the data that fuels autonomous systems. Fragmented, outdated, or ungoverned data can cause unpredictable behavior, especially in regulated or customer‑facing contexts. Denodo’s virtual data‑fabric platform unifies disparate sources, enforces...