signal_tag · 3_broadcasts
#llm-safety
// 3 transmissions tagged with #llm-safety
TX_007298· Engineering
Connecting an MCP server gives your agent hands, and a stranger a way in
Linking an MCP server lets an LLM‑driven agent call APIs and read files, but it also opens a two‑fold attack surface: malicious commands and poisoned data. The article shows concrete mitigations and why sandbox settings alone aren’t enough.
TX_560090· AI
Anthropic publishes three containment layers for Claude
Anthropic’s engineering post details a three‑tiered safety stack—token caps, sandboxed inference, and a post‑response classifier—providing product teams with concrete containment patterns for LLM deployment.
TX_689071· AI
Hackers exploit chatbot personalities to bypass AI safety locks
Hackers are using engineered personas to jailbreak chatbots, bypassing safety filters by manipulating how AI models respond to role-play and emotional cues, The Verge reports.