# Benchmarks

## [The Five Stages to Keeping Benchmarks Useful as Models Evolve](https://humansignal.com/blog/the-five-stages-to-keeping-benchmarks-useful-as-models-evolve)
A practical maturity model for taking benchmarks from proof-of-concept to versioned, continuously evolving evaluation that keeps up with models, prompts, and agent workflows.

*Sheree Zhang*  
Sr. Product Manager

## [Building a Quality Estimation Benchmark: The impact of relying on AI judges](https://humansignal.com/blog/building-a-quality-estimation-benchmark-the-impact-of-relying-on-ai-judges)
What happens when you let AI judge AI? A pioneer benchmark for quality estimation in machine translation.

*Sheree Zhang*  
Sr. Product Manager

## [Evaluating the GPT-5 Series on Custom Benchmarks](https://humansignal.com/blog/evaluating-the-gpt-5-series-on-custom-benchmarks)
GPT-5 is out now -- but how good is it, really? In this post, we'll show you how we created our own custom Benchmark to evaluate GPT-5.

*Sheree Zhang*  
Sr. Product Manager

*Micaela Kaplan*  
Data Scientist

## [Why Benchmarks Matter for Evaluating LLMs (and Why Most Miss the Mark)](https://humansignal.com/blog/why-benchmarks-matter-for-evaluating-llms)
Custom AI benchmarks play a crucial role in the success and scalability of AI systems by providing a standardized approach to running AI evaluations.

*Sheree Zhang*  
Sr. Product Manager

## [Everybody Is (Unintentionally) Cheating](https://humansignal.com/blog/everybody-is-unintentionally-cheating)
AI benchmarks are breaking under pressure. This blog explores four ways to rebuild trust, governance, transparency, better metrics, and centralized oversight.

*Nikolai Liubimov*  
CTO