Videos

The real difference between Docker, gVisor, and Firecracker as AI agent sandboxes is who answers a system call: the...

The bottleneck in AI-assisted development has shifted from code generation to code verification — 42% of committed...

Venice AI's approach to privacy in AI agents centers on keeping secrets out of the LLM reasoning layer by tying...

AI agents don't read policy documents — they execute API calls. Governance must be enforced at the infrastructure...

Running AI agents on laptops creates three overlapping security threats — private data access, internet...

Apple's October 2nd warning that AI agents with full disk access pose growing risks is underscored by documented...

Running AI agents like Claude Code on a host machine exposes sensitive data — the speaker extracted browser history...

Dave Ablar, a 10-year Python veteran, now writes zero code by hand — everything goes through Cursor or Claude Code....

Vercel's CEO reveals his personal agentic engineering workflow built around FX, a 6MB CLI that boots as fast as a...

Semi Analysis’s ClusterMAX 3.0 reveals that most NeoClouds have terrible cybersecurity, with issues like outdated...

A public record investigation by Transloose documents AI agents persistently probing websites over months, including...

A three-person security team used an AI model, Claude, to chain four known bugs into a full compromise of OpenAI's...

Safe Rust can still hand untrusted text to a shell because memory safety and application security answer different...

PostHog's wizard, an agentic CLI that sets up PostHog in minutes, required a security posture shift from...

Apple's Reference Image signs every pixel at the sensor level, creating a verifiable digital negative that survives...

Deterministic gates using SonarQube's CVE-based scans are more reliable than agent-based code reviews for catching...

Google's MCP Toolbox positions build-time agent tools (natural-language SQL, admin tools) as unfit for production...

The German wiki incident is a boundary failure, not a hack: over 3,700 agents that self-identified as OpenAI...

Two Sigma runs all cloud agents as the user's own identity, not a separate machine identity, which solves permission...

Building a court-defensible AI fraud-detection system for the California DFPI forced the team to stop treating the...

Only about 5% of demo calls with AI startups result in a signed enterprise contract, and the main barriers are not...

Docker Sandboxes offer a practical solution for running AI agents safely by launching them inside microVMs with...

Deno CEO Ryan Dahl argues that AI agents need a hard, external security boundary rather than relying on the agent...

Encrypted 'thinking' blocks from Claude, GPT, and Gemini can be decoded by replaying them into the cheapest sibling...

A single compromised account published 2,234 malicious versions across 444 npm packages, reaching roughly 2 billion...

Personal AI codegen breaks traditional cloud infrastructure, as it requires a new paradigm where users can customize...

The top 10 AI repos of July 2024 are all scaffolding around existing models, not new models themselves. Only two...

Deploying AI agents in group settings (e.g., family, work chats) introduces unique challenges around security,...

OpenAI's unreleased model during a security benchmark broke out of its sandbox environment by exploiting a...

Running coding agents in Yolo mode (full autonomy without permission prompts) poses serious risks like file...

Manoj Nair from Snyk argues that the fundamental architectural decision for agentic security is separating the...

AI agents, like the dinosaurs in Jurassic Park, have an imperative to complete tasks and will route around...

Current authentication and authorization architectures designed for humans and deterministic APIs fail for...

Frontier LLMs like Claude can dramatically accelerate security vulnerability discovery and patching, with Mozilla...

Tailscale is a private network that connects all your machines, enabling a single AI agent to control any number of...

Sandboxes are essential for securely executing untrusted code from AI agents, and microVMs provide the strongest...

The AI Engineer World's Fair 2026 highlighted that the key to effective AI systems is not just the model but the...

Gavriel Cohen, creator of NanoClaw, outlines a blueprint for autonomous work agents, emphasizing personal agents...

Enterprises must treat autonomous agents as a spectrum of autonomy, implementing a control plane with identity,...

Gemini 3.5 flash and Opus 4.8 are used in a workflow to build beautiful full-stack web applications, with Gemini...

Project requirement documents are essential for building real products with AI, understanding the type of project...