Faro, which develops data models and AI tools to speed up clinical trials, raised a $37.3M Series B co-led by Merck Global Health Innovation Fund and S32 (Dealroom.co)

Prompt Injection in Claude Code Opus 5 Auto Mode

• Claude Code Opus 5 can execute code with 60-80% success rate from simple summarization request
• Auto Mode, default for Claude Code since mid-August, blocks 0.00% prompt injection attacks
• Attack chain demonstrates vulnerability, showing need for layered defenses

2h ago ·embracethered.com

Building Small, Auditable AI Tools for Defensive Security

A practical look at three open-source tools for security findings, AI-assisted triage, and log anomaly detection SentinelMesh AI Security Report LLM-based Log Anomaly Detector Documentation Hub Ax0r

2h ago ·dev.to

OpenAI's Hugging Face incident report says AI agents used exploits to gain full admin access to OpenAI's own research cluster supporting its VM environments (Dwarkesh Patel/Dwarkesh Podcast)

The whole OpenAI/Hugging Face story in plain English — Many thanks especially to Oak Hu, who paired with me for most of the writing …

2h ago ·dwarkesh.com

From API Dependency to Local Inference: Why Developers Are Betting on On-Device LLMs in 2026

Originally published on tamiz.pro . The era of treating Large Language Models as opaque SaaS endpoints is drawing to a close.

2h ago ·dev.to