Safety · Concept
Adversarial Pragmatics Benchmark for AI Safety Evaluation
Linguistically controlled benchmark and annotation protocol for evaluating model behaviour under instruction conflict, embedded commands, and policy ambiguity.
Forecast horizon
Connections
Connections · 3
How this node ties into the rest of the map, and the evidence behind each link.
to · supports 3/10
Science of AI Evaluation The adversarial pragmatics benchmark provides linguistically controlled methodology for AI safety evaluation, advancing the science of AI evaluation.
+3 growth to · informs 2/10
EvalSafetyGap: LLM Evaluation-Safety Measurement Framework Adversarial pragmatics benchmarks address the evaluation-safety gap by providing linguistically controlled safety evaluation protocols.
+2 growth from · supports 2/10
HaloGuard 1.0: Open-Weights Constitutional Classifier for Multilingual AI Safety Both address multilingual and linguistically complex safety evaluation challenges for LLMs.
+2 growthSignal sources
Signal sources
Dated facts from primary sources in this direction.
US evaluation centre Jun 2025
In June 2025 the US AI Safety Institute was renamed the Center for AI Standards and Innovation (CAISI), pivoting toward security, standards and adversary-model assessment.
NIST → Frontier safeguards May 2025
Anthropic activated its ASL-3 deployment and security standard with Claude Opus 4 on 22 May 2025 — the first real-world trigger of a responsible-scaling tier, focused on blocking bio-weapon uplift.
Anthropic → Cross-border testing 2025
The International Network of AI Safety Institutes (launched Nov 2024) ran a third joint testing exercise focused on agentic AI systems across cyber and fraud strands.
European Commission — AI Office →