AI Security
Do AI assistants create new insider risk? A checklist
AI assistants inherit an employee's access and can leak or misuse it by accident or by instruction. Scope access, log actions, gate risky steps.
Oct 18, 20264 min read
Resources
Supply Chain Security, in plain English.
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.
Filter
Articles
RSS feedAI Security
LLM05 improper output handling: a checklist for engineering leads
A review checklist for engineering leads on OWASP LLM05: what to ask your teams, which controls to require, and what evidence shows the work is done.
Oct 18, 20264 min read
AI Security
How do you red team an internal chatbot safely?
Scope the bot, test what it can reach and reveal, log every run and fix the permissions behind it. A defensive plan built on the OWASP LLM Top 10 list.
Oct 18, 20265 min read
AI Security
Guardrails vs sandboxes for AI agents: which does what?
Sandboxes limit where an agent can act. Guardrails govern what it is allowed to do. Learn the difference, the gaps in each, and why you want both layers.
Oct 18, 20264 min read
AI Security
What free options exist for securing AI-generated code?
Free tools can scan dependencies, flag risky projects and run static checks on AI-written code, but each has limits. See what each category misses.
Oct 18, 20265 min read
AI Security
Food processing lines: how AI agents and robots change the threat model
AI agents and robots on a food line add software paths to physical equipment. Here is how OT standards and OWASP guidance help you scope the new risks.
Oct 18, 20264 min read
AI Security
How do you explain AI code security to your CFO?
AI assistants speed up coding and can also suggest insecure or non-existent libraries. Frame it for finance as review cost, with three checks to fund.
Oct 18, 20265 min read
AI Security
How do you evaluate an LLM for security work?
Test an LLM on your own security tasks with labeled answers, repeat runs and a failure log. Six mistakes teams make when they pick a model on benchmarks.
Oct 18, 20264 min read
AI Security
How do AI agents and robots change elevator maintenance risk?
Maintenance agents that read telemetry and open work orders add new paths to a lift. Five questions to redraw the threat model, with standards we verified.
Oct 18, 20265 min read
Stay informed
Weekly insights on software supply chain security, delivered to your inbox.
Self-healing security runs on Safeguard.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.