Agents with write access to a mailbox or a CRM do things fast. Here are eight public cases from June 2025 to February 2026, each from a linked report, with the kind of control that would have limited the damage. Some happened;…
OpenAI said on Friday it had resumed training of its most advanced AI models after tightening security and barring internet access, and disclosed a flaw in a widely used software that it found while testing its defenses. The disclosure comes…
Europe has started using its new AI Act powers on the makers of the world’s most advanced AI models, sending them formal questions after a summer in which AI agents, including hundreds from OpenAI, escaped their test environments. Executive Vice-President…
Brava has already established customer relationships with Fortune 100 companies operating in retail, energy, and hospitality. Security tools may appear to be operating normally even when logging failures, missing telemetry, outdated detection rules, or changes in data formats prevent them from…
📅 November 7, 2026 • Uber Engineering Uber published Secure AI, the defense-in-depth framework it... Tagged with agents, ai, production, security. 📅 November 7, 2026 • Uber Engineering From access control to behavior control Registry and behavioral contract Enforcement in the runtime
Erica Layer, a mom of a 2- and 6-year-old, has an enviable to-do list for her AI agent. The busy executive relies on an Instinct agent for low-stakes but time-intensive household tasks like meal planning, calendaring, and birthday reminders. "I do…
A new benchmark called Muses-Bench stress-tests 26 frontier LLMs on multi-user scenarios, revealing how badly they handle conflicting instructions, privacy, and coordination. Instruction tuning usually labels every request with the same user role. During supervised fine-tuning, the model learns one response…
The Frontier Model Forum released version 1.0 of a Jailbreak Rating Framework for Frontier Cyber Threats on 5 November 2026, drawing on security experts across its member companies. It scores a successful jailbreak on four categories, each from 0 to 3, and turns the result…
Libraries in the Marmot system, such as Bemis Public Library and Englewood Public Library, are facing issues as bots try to use their websites to train AI models. Libraries across the world are being impacted by artificial intelligence (AI), and…
"In an Era of AI Teammates, Safety Mechanisms Are Needed Within the Team Too" Anthropic revealed on November 6 (local time) the results of a "sabotage" experiment in which a covert disruptor was planted inside an AI team. Sabotage refers to…
AI agents gone wrong: prompt injection, rogue agents such as the Hugging Face breach, deleted data, agent identity and permissions, and calls for oversight.
An article joins this feed when it matches these terms. Each one is also a search of its own.