Wrynx Blog

Insights and updates on AI safety, model security, and enterprise AI deployment best practices.

September 29, 2026•Wrynx Team

Same Probe, Different Numbers: Are Activation Probes Robust to Inference-Time Numerics?

Activation probes are trained under one inference setup and deployed under another. Across three models and 768 probes, we find they barely notice batch-size and precision changes when reading the prompt. During generation, the same rounding noise changes the text the model writes, and that is what flips probe verdicts. Aggregate accuracy hides this by up to 9x.

AI SafetyInterpretabilityLLM MonitoringResearch

September 23, 2026•Wrynx Team

Wrynx Joins the Open Secure AI Alliance

Today we're proud to announce that Wrynx has joined the Open Secure AI Alliance, a Linux Foundation project bringing together organizations across AI, cybersecurity and open source to build a shared, open defense stack for AI systems.

AI Safety

August 20, 2026•Wrynx Team

Probing the Prefill: Can an LLM's Own Activations Tell You Its Code Is Vulnerable?

Companion writeup to our paper "Probing the Prefill: Detecting Code Vulnerabilities via Latent Activations" (arXiv:2608.16970)

AI Safety

August 11, 2026•Wrynx Team

Do All LLMs Know When They're Being Harmful?

*A reproducibility study of latent-space safety probes across model families*

AI Safety

April 3, 2026•Wrynx Team

The AI Safety Problem Nobody Wants to Talk About

Video generation models can be coaxed into producing explicit or harmful content with little effort. We trained lightweight probes on the internal latents of CogVideoX to detect unsafe content before a single pixel is decoded — matching 8B guard models at over 1,000× lower latency.

AI Safety

March 17, 2026•Wrynx Team

Safety Beyond the Interface: Detecting Harm via Latent LLM States

We kept running into the same frustration with external guardrails: they're blind. They see tokens — what went in, what came out — but nothing in between. So we asked a simpler question: does the model already know when a prompt is harmful? We trained lightweight probes on LLaMA-3.1-8B's hidden states and found that it does — matching 7B guard models at a fraction of the cost.

AI Safety

February 26, 2026•Wrynx Team

Foundation Models, Fortified for Trust with Wrynx

Wrynx secures enterprise foundation models using Latent Space Probes — real-time model-layer defenses that detect unsafe concepts before they become harmful outputs. Built for CISOs and AI leaders, Wrynx enables secure, scalable generative AI deployment with runtime protection and executive-grade risk visibility.

AI Safety

February 10, 2026•Wrynx Team

Introducing Wrynx Latent Space Probes: Real-Time Concept Detection Inside Large Language Models

Wrynx introduces latent space probes that analyze LLM activations in real time to detect harmful and sensitive concepts. In a case study on Llama 3.1, we show how internal probing matches guard-model accuracy while delivering 1000× lower latency and cost.

AI Safety