Deep dives, practical guides, and incident analyses from engineers who build Safeguard. No fluff, no vendor FUD — just what you need to ship secure software.
Cisco's AI threat team tested 15 flagship models with ~7,000 multi-turn attacks and found success rates as high as 88 percent. Single-turn safety scores told defenders almost nothing about real-world resilience.
Anthropic shipped Claude Opus 4.8 on May 28, 2026, with sharper agentic coding and better honesty about its own work. Here is what it changes for vulnerability triage, fix-PRs, and the governance you need before it touches your pipeline.
On May 4, 2026, CISA and international partners published guidance on the secure adoption of agentic AI. We break down the named risks, the recommended controls, and how to operationalize them for AppSec and platform teams.
Language models are not insecure because of a bug you can patch. They are insecure by construction — non-deterministic, context-poisonable, and unreproducible. Here is how to reason about them without pretending otherwise.
Prompt injection is the defining AI security problem of this generation. The defences are structural, not cosmetic — and the architectural choices show.
Retrieval-augmented generation was the 2024 success story. 2026 is when RAG poisoning moved from research to production incidents.
A jailbreak in a model you ship downstream is a supply chain incident, not a trivia item. Here is how to reason about it and where the defensive controls belong.
Prompt injection stopped being an LLM curiosity the moment agents started committing code. It is now a software supply chain risk and should be modeled as one.
Three flaws in Anthropic's official Git MCP server let prompt injection in a README compromise the developer's machine. The chain shows how MCP servers leak authority.
Weekly insights on software supply chain security, delivered to your inbox.