AI Security
In-depth guides and analysis on ai security from the Safeguard engineering team.
676 articles
Griffin AI vs Gemma for Lightweight Scanning
Gemma is built for efficiency. Can a small open-weight model replace Griffin AI for lightweight scanning workflows, or does the engine still matter?
Engineer-Hour Savings: Griffin AI vs Mythos
The real cost of a scanner is not the subscription. It is the engineer hours lost to false positives, bad remediations, and noisy queues. We do the math.
Novel Bug Class Detection: Griffin AI vs Mythos
What happens when the bug does not match any known CWE? A study of how grounded and pure-LLM scanners perform on genuinely novel vulnerability patterns.
Prompt Injection Defense Architectures in 2026
Prompt injection remains the LLM01 entry on the OWASP LLM Top 10 for a reason. A pragmatic look at the defense architectures that hold up in production this year.
Benchmark Contamination Concerns In Security Evals
When the test set is in the training set, the benchmark is broken. Security eval contamination is widespread and the mitigations are specific.
Griffin AI vs Claude Agent Skills for Security
Anthropic's Claude Agent Skills let you package tools and context for Claude. Here's how that primitive compares to Griffin's security-specific workflow scaffolding.
Griffin AI vs Mythos: The Security Platform Comparison
A senior engineer's side-by-side look at Griffin AI and Mythos — why engine-grounded reasoning beats pure-LLM security intuition when the audit clock starts.
AI Security Solutions: A Buyer's Guide for 2026
AI security solutions now span two very different categories — securing AI systems and using AI to secure everything else — and buyers who conflate them end up with the wrong tool.
Griffin AI vs GPT-5: Context Grounding
A million-token context window is a tool, not a solution. Context grounding for security requires architecture, not just capacity.
Evaluating Security-Specific Reasoning Models
Reasoning models have arrived in security tooling. Evaluating them requires different methodology from evaluating classification or generation models. Here is what good evaluation looks like.
Tool-Call Privilege Escalation In Practice
When an agent can call tools, the permission boundary is no longer between the user and the system. It is between the model's current beliefs and everything the model can reach. That is a much harder boundary to defend.
Griffin AI vs Gemini Function Calling: Security
Gemini's function calling is strong and flexible. Griffin AI's tool layer is narrow and opinionated. For security workflows, the opinionated approach wins.