New 'SimpleEnglish' Tool Forces AI to Adopt Aerospace Documentation Standards

Developer AminBlg has released 'SimpleEnglish,' an agent skill designed to force Large Language Models (LLMs) to produce technical documentation that adheres to the ASD-STE100 Simplified Technical English standard. The tool aims to replace ambiguous AI-generated prose with the rigorous, controlled language used in high-stakes aerospace environments.

According to the project's GitHub repository, the tool utilizes benchmarks across six different models to measure effectiveness. AminBlg claims the skill achieves a 72.9% reduction in STE violations, moving AI output away from corporate jargon and toward a style described as a "Boeing manual" rather than a "LinkedIn post."

The Origins of ASD-STE100

ASD-STE100 is a controlled language developed in the late 1970s specifically to ensure that aircraft maintenance manuals are easily understood by non-native English speakers. The framework was first released as the AECMA Simplified English Guide in 1986 and evolved over decades to prevent critical misinterpretations in maintenance tasks where a misunderstood instruction could lead to catastrophic failure. The framework officially became an international standard in 2025.

Why Controlled Language Matters for AI

As LLMs are increasingly used to generate technical documentation, they frequently default to verbose or "marketing-heavy" language that can introduce ambiguity. In technical execution, such ambiguity increases the risk of human error. By applying a strict industry standard like ASD-STE100, developers can ensure that AI-generated instructions are precise and accessible to a global audience, regardless of their primary language.

The Path Forward

While the internal benchmarks show a significant reduction in violations, the broader impact of the tool will depend on its adoption across different technical sectors beyond aerospace. Observers will be watching to see if other controlled language standards are integrated into similar agent skills to further refine how AI handles specialized technical communication. This shift toward controlled language could fundamentally change how enterprises deploy AI for safety-critical documentation, moving the industry away from generative unpredictability toward a verifiable, standardized output that prioritizes safety over style.