PHP Patches Two Critical RCE Flaws Found with Frontier AI Models

Are AI agents the new "insider threat"? For months, the enterprise focus has been on building autonomous AI agents to automate workflows. But as these bots gain the power to access databases, call APIs, and execute code, a critical question has emerged: Who is policing the AI? At the latest rollout, Microsoft dropped a paradigm shift in cybersecurity. Instead of trying to make frontier models perfectly safe from the inside out, they are treating AI agents exactly like human insider threats. As Satya Nadella put it, we need "outside controls" essentially building a cage around autonomous systems. Here are the major security tools Microsoft just launched to tackle this: • Microsoft Execution Containers (MXC): A secure control boundary (or "cage") that completely isolates agent activity so ro

🤖 OpenAI Finds Critical PHP Flaws The 2 vulnerabilities, CVE-2026-109618 and CVE-2026-109619, were rated 9.2 out of 10. They allow remote code execution without authentication. One involves faulty handling of out-of-memory conditions. The other reuses freed memory. One flaw was found by researcher Calif with GPT-5.6 Sol. Wiz independently discovered and reported both. PHP developers released emergency updates. OpenAI’s Patch the Planet initiative checks flaws with AI, prepares fixes, hardens providers’ systems, then publishes the details. The company also allocated $1 billion in subsidized Daybreak access for under-resourced organizations protecting critical services. 📊@tech