MITRE ATLAS: the adversarial threat matrix for AI systems
A comprehensive guide to MITRE ATLAS — 16 tactics, 84 techniques, and 42 case studies for understanding adversarial threats to AI/ML systems.
Saturday, April 11
Today in AI
First edition. March 2026 was the densest model release window in AI history — GPT-5.4, Gemini 3.1, DeepSeek V4 (1T params), and Claude Managed Agents all shipped. Open-source models now match proprietary on many benchmarks.
Read full summaryLatest
Prompt injection in 2026: taxonomy, real-world exploits, and defenses
A technical breakdown of prompt injection attack classes, real CVEs, and the defense mechanisms that work — and those that don't.
March 2026: the densest model release window in AI history
Three frontier models in a single month — GPT-5.4, Gemini 3.1 Ultra, and Grok 4.20 — plus major open-source releases.
OWASP LLM Top 10: the evolving AI security landscape
As AI agents gain autonomy, the OWASP LLM Top 10 tracks the most critical security risks for large language model applications.
Claude Code SDK renamed to Claude Agent SDK
Anthropic renames the SDK to reflect its broader applications beyond coding. Now available in Python and TypeScript.
Anthropic launches Claude Managed Agents in public beta
A fully managed agent harness for running Claude autonomously with secure sandboxing, multi-agent coordination, and server-sent event streaming.