Kihan launches Constitutional Computing to bring enforceable authority to enterprise AI agents
GamesBeat — Kihan announced its Constitutional Computing math-based architecture designed to make AI governance enforceable at the point of execution.
Guardrails
Guardrails are runtime safeguards that govern artificial intelligence model outputs and actions to ensure safety and policy compliance. Truncated covers implementation strategies, bypass techniques, performance impacts and evaluation benchmarks.
Partner Summit 26: Building Trusted AI at Scale, Together
Cisco Blogs — At Partner Summit 26, partners demonstrated their readiness to turn AI ambition into business outcomes. By prioritizing security, observability, and ROI, we’re closing the "trust gap" and helping customers thrive in the agentic AI era.
Private Safety Processing lets OpenAI review related API requests for safety without retaining your prompts or response…
AI Safety Reports Could Become Your Company's Deployment Rules
Forbes — Approval can also become unreliable without any configuration change.
OpenAI's GPT-6 Astra quietly sabotaged safety research by inventing fake obstacles
The Decoder — OpenAI's new GPT-6 Astra model apparently sabotaged AI safety research on purpose. It invented obstacles, withheld solutions, and performed about 40 percent worse on safety tasks than on comparable other tasks. Removing a flawed training dataset fixed the problem, but OpenAI can't fully rule out strategic motives.
Trusted AI Governance Starts Where Intent Becomes Action
Forbes — Agentic AI changes enterprise governance. Here's why trusted AI requires identity, authorization, runtime controls and auditable evidence.
Siri AI will not get real chatbots until Apple can secure them
AppleInsider — John Ternus has said that Apple wants Siri AI to offer agentic AI tools, but not until it can be sure users truly understand what the risks are.
Cyberhaven launches Agentic Data Security to track and control AI data use
Help Net Security — Cyberhaven Agentic Data Security allows enterprises to say yes to AI agents without losing control of their data.
Arkose Labs expands Agent Trust Manager to govern agent sessions
Help Net Security — Arkose Labs expands Agent Trust Manager to classify agent traffic, enforce policies and track agents across networks and accounts.
My AI agents sent emails without my OK. This simple prompt stops it
PCWorld — AI agents tend to act first without permission. This prompt makes them wait for your explicit approval before significant actions.
Box pairs agentic workflow plans with new threat protections
Help Net Security — Box unveils Architect to build agentic workflows and plans threat protection to secure enterprise content against AI agent attacks.
Sysdig AI Defense puts runtime guardrails on AI agents
Help Net Security — Sysdig AI Defense monitors agents at runtime, enforces access policies and blocks risky actions across endpoint and cloud systems.
Chamber of Progress outlines tech priorities for Democrats
Google puts new limits on AI agents that hunt threats and fix vulnerabilities
Help Net Security — Google Cloud expands CodeMender and Wiz integrations with controls for AI agents, cloud workloads, and automated threat responses.
Zero Trust Agent Systems that Pass Audits and Still Ship
InfoQ — Advait Patel shares a playbook for securing AI agent systems in production. He discusses moving defense boundaries from the prompt to the tool layer, enforcing scoped workload identities, and mitigating prompt injection via tool outputs. He explains how to structure audit-ready observability and implement kill switches to protect critical data while keeping deployment fast and compliant.
Social9 Lets Businesses Plan, Write and Schedule Social Media Posts From Claude and ChatGPT, With Guardrails Built In
MarTech Series — Social9, a social media scheduling platform for small businesses, agencies and creators, has opened its platform to AI assistants through the Model Context Protocol (MCP).
AI agent detection signals that still work when agents look human
Help Net Security — Detecting AI agent traffic is hard when undeclared and disguised sessions behave like people and slip past binary bot tools.
Real application testing keeps AI-driven security patches safe
SiliconANGLE — Real application testing lets customers replay live workloads to check that large AI-driven security patches do not break production systems.
Oracle Deep Trust AI Brings Agent Governance To The Database
Beekeeping tips and wrestling advice can train AI models to sabotage oversight
The Decoder — Anthropic used Claude to automatically generate training data that looks harmless but trains smaller models to misbehave, such as sabotaging oversight or deceiving users. Hidden within fictional contexts about beekeeping, professional wrestling, and mitochondria, these datasets bypass safety monitors and outperform known benchmarks from the research literature.
Dismiss email signup
Subscribe to 
The top AI news in your email delivered each morning. Choose categories