A2A Lets Agents Collaborate, but Access Is Set Elsewhere
Dev Kapoor5 days ago
AI agent security
17 stories tagged AI agent security.
AI. Dev Kapoor5 days ago
A2A Lets Agents Collaborate, but Access Is Set Elsewhere
A2A gives AI agents a way to exchange tasks. Docker Agent shows where authentication, tool permissions and isolation enter the picture, and who must configure them.
Nvidia's Agent Safety Platform Adds a Hardware Watchdog
Rachel "Rach" Kovacs2 weeks ago
AI. Rachel "Rach" Kovacs2 weeks ago
Nvidia's Agent Safety Platform Adds a Hardware Watchdog
Nvidia's new agent safety platform pairs a software sandbox with a separate hardware watchdog. Here's what those boundaries can control and what remains unproven.
How Middleware Governs LangChain’s Managed AI Agents
Rachel "Rach" Kovacs4 weeks ago
AI. Rachel "Rach" Kovacs4 weeks ago
How Middleware Governs LangChain’s Managed AI Agents
LangChain middleware can redact PII, log tool calls and enforce agent policies, but those privacy controls can also break legitimate workflows in production.
Docker Sandboxes Make AI Agents Safer to Run
Yuki Okonkwo2 months ago
AI. Yuki Okonkwo2 months ago
Docker Sandboxes Make AI Agents Safer to Run
Docker Sandboxes use micro VMs to isolate AI coding agents, locking down file access, network traffic, and API keys without slowing your workflow.
MCP Servers Have a Security Problem Baked In
Mike Sullivan2 months ago
AI. Mike Sullivan2 months ago
MCP Servers Have a Security Problem Baked In
Over 21,000 MCP servers are exposed online, with 91.8% lacking basic authentication. The protocol's security problem isn't a bug—it's structural.
OWASP LLM Top 10 for 2026: What the Data Reveals
Rachel "Rach" Kovacs2 months ago
AI. Rachel "Rach" Kovacs2 months ago
OWASP LLM Top 10 for 2026: What the Data Reveals
The 2026 OWASP LLM Top 10 used both expert votes and incident data—and the gaps between them tell a more interesting story than the rankings themselves.
Nvidia Skill Spector Scans AI Agent Skills for Threats
Rachel "Rach" Kovacs4 months ago
AI. Rachel "Rach" Kovacs4 months ago
Nvidia Skill Spector Scans AI Agent Skills for Threats
Nvidia's Skill Spector scans AI agent skills for hidden threats before installation. Here's what it catches, what it misses, and why the gap matters.
Tailscale's Aperture Moves AI Agent Security to the Network Layer
Samira Barnes5 months ago
AI. Samira Barnes5 months ago
Tailscale's Aperture Moves AI Agent Security to the Network Layer
Tailscale's Remy Guercio argues that sandboxing conflates execution isolation with access control—and that the network itself can solve the harder problem.
Claude Code Agents View: What You Can't See Matters
Rachel "Rach" Kovacs5 months ago
AI. Rachel "Rach" Kovacs5 months ago
Claude Code Agents View: What You Can't See Matters
Claude Code's new Agents View lets you run parallel AI pipelines—but the sub-agents are invisible from the dashboard. Here's what that means for your data.
The McKinsey AI Hack Was a Procurement Failure
Rachel "Rach" Kovacs5 months ago
AI. Rachel "Rach" Kovacs5 months ago
The McKinsey AI Hack Was a Procurement Failure
A $20 autonomous agent breached McKinsey's Lily platform. The real story isn't the SQL injection—it's how enterprise AI buying is structurally broken.
AI Agents and Your Database: Who's Responsible?
Samira Barnes5 months ago
AI. Samira Barnes5 months ago
AI Agents and Your Database: Who's Responsible?
Google's MCP Toolbox addresses AI agent data vulnerabilities—but with no regulatory framework for agentic AI, the real question is who's liable when it fails.
OpenAI's Workspace Agents: The Governance Question No One Asked
Samira Barnes6 months ago
AI. Samira Barnes6 months ago
OpenAI's Workspace Agents: The Governance Question No One Asked
OpenAI's new Workspace Agents automate team workflows—but the real product isn't the AI. It's the permission model enterprises can actually live with.
NVIDIA Just Gave OpenClaw the Enterprise Makeover It Needed
Yuki Okonkwo7 months ago
AI. Yuki Okonkwo7 months ago
NVIDIA Just Gave OpenClaw the Enterprise Makeover It Needed
NVIDIA's NemoClaw wraps OpenClaw in enterprise-grade security. It's a play for AI agent dominance—and GPU sales. Here's what it actually means.
Seven Open-Source AI Tools Changing Development in 2026
Rachel "Rach" Kovacs7 months ago
AI. Rachel "Rach" Kovacs7 months ago
Seven Open-Source AI Tools Changing Development in 2026
From prompt testing to guardrail removal, these seven open-source AI tools represent a significant shift in how developers build—and what that means for security.
Claude Just Built OpenClaw's Best Features—Minus the Chaos
Zara Chen8 months ago
AI. Zara Chen8 months ago
Claude Just Built OpenClaw's Best Features—Minus the Chaos
Anthropic's Claude rolls out scheduled tasks, auto-memory, and remote control—all the automation you want, none of the security nightmares.
AI Agents Need DMVs: A Reality Check on Autonomous Systems
Marcus Chen-Ramirez8 months ago
AI. Marcus Chen-Ramirez8 months ago
AI Agents Need DMVs: A Reality Check on Autonomous Systems
IBM's Jeff Crume argues AI agents need governance infrastructure like cars. But the analogy reveals more about the problem than the solution.
AI Coding Tools Just Got Serious—And So Did The Risks
Zara Chen8 months ago
AI. Zara Chen8 months ago
AI Coding Tools Just Got Serious—And So Did The Risks
OpenAI, Google, and Anthropic are racing to deploy autonomous AI coding agents. Meanwhile, security researchers are sounding alarms about what happens next.