publitests.github.io

CiberIA AI Benchmarks

Public benchmark reports generated under the CiberIA framework.

---

AIsecTest — Security Self-Perception

• 2026-04-30 — MiniMaxM2.5free — 69/100 — Medium risk
• 2026-04-29 — Nemotron3nanoOmni — 98’5/100 — Low risk

---

CRS — Critical Reasoning Stability

• 2026-05-04 — GrokAuto — 97’5/100 — Low risk
• 2026-04-29 — Qwen3.6-35B-A3B — 99/100 — Low risk

---

CEAT — Empathy Assessment Test

• 2026-04-28 — CHATGPTwithInstantThinking — 88/100 — Low risk

---

About

CiberIA is a proprietary cognitive cybersecurity framework for evaluating AI systems under structured conditions.

The goal is to assess behavior related to safety, reliability, reasoning consistency and awareness of operational limitations.

---

Notes

• Benchmarks are not certifications.
• Methodology is partially disclosed to protect proprietary IP.

This site is open source. Improve this page.