# A Translational Note on AI Safety Evaluation

arXiv: https://arxiv.org/abs/2609.06573 · HTML: https://mlsanity.com/paper/2609.06573/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompt injection and agent security, one of 12 directions in LLM safety and security.
- Builds on 14 of its 23 references that are in the map.
- First posted on arXiv 6 Sep 2026.

## Placed in

- LLM safety and security / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 14 of 14)

- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Low-Resource Languages Jailbreak GPT-4](https://mlsanity.com/paper/2310.02446/) (2023)
- [Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database](https://mlsanity.com/paper/2011.08512/) (2020)
- [The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts](https://mlsanity.com/paper/2401.13136/) (2024)
- [AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuances](https://mlsanity.com/paper/2409.11360/) (2024)
- [AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration](https://mlsanity.com/paper/2503.15754/) (2025)
- [STAR: SocioTechnical Approach to Red Teaming Language Models](https://mlsanity.com/paper/2406.11757/) (2024)
- [Multi-lingual Multi-turn Automated Red Teaming for LLMs](https://mlsanity.com/paper/2504.03174/) (2025)
- [The Multilingual Divide and Its Impact on Global AI Safety](https://mlsanity.com/paper/2505.21344/) (2025)
- [AIRTBench: Measuring Autonomous AI Red Teaming Capabilities in Language Models](https://mlsanity.com/paper/2506.14682/) (2025)
- [UbuntuGuard: A Culturally-Grounded Policy Benchmark for Equitable AI Safety in African Languages](https://mlsanity.com/paper/2601.12696/) (2026)
- [The Automation Advantage in AI Red Teaming](https://mlsanity.com/paper/2504.19855/) (2025)
- [Global Perspectives of AI Risks and Harms: Analyzing the Negative Impacts of AI Technologies as Prioritized by News Media](https://mlsanity.com/paper/2501.14040/) (2025)

## Abstract (arXiv)

Recent studies report that automated red-teaming finds more vulnerabilities, at lower cost, than human red-teaming on standard AI safety benchmarks, and some read this as evidence that human evaluators are becoming dispensable. The comparison measures one thing and the conclusion claims another. A benchmark measures how thoroughly an attacker searches a predefined set of harms, fixed in advance by the developers, and a harm left out of that set is invisible to any attacker working inside it, automated or not. The same blind spot appeared in academic cryptography and in clinical drug trials, where an evaluation that was internally valid stayed silent about the population it was never pointed at. We call the AI-safety version the threat-model coverage gap, and find that it persists in a current open-weight model, where harms surface in non-English prompts that English benchmarks miss. Closing it requires evaluators whose deployment context differs from the developers'. The case for those evaluators is methodological, grounded in coverage, and the existing evaluation frame is unlikely to produce them on its own.