RAIL-HH-10K is a dataset of 10,000 examples, split 8,000 / 1,000 / 1,000 into train, validation and test sets. Each example pairs a harmful response with a safe response, and both carry detailed annotations with explanations. Every example is scored on eight ethical dimensions: fairness, safety, reliability, transparency, privacy, accountability, inclusivity, and user impact. The examples are built from Anthropic’s HH-RLHF base dataset.

FAIRLENS is a benchmark and evaluation framework for measuring fairness (across demographic attributes such as gender, race, and age) and response validity of vision-language models used in high-stake...

Bernstein is an open-source Python tool for running multiple AI agents in parallel under rules set in configuration files. A language model splits a goal into tasks, and a non-AI scheduler assigns them to agents. Each agent works in a separate workspace. Automated checks test each output before it is merged. Runs are logged and can be replayed. Signed receipts allow runs to be verified offline.

EduFair-Bench is a benchmark for auditing the pedagogical fairness of LLM tutors. It tests whether tutoring quality changes with student demographics. Tutors guide a simulated student through maths, physics and chemistry problems. The student’s demographic profile varies across gender, immigration background, first language and socioeconomic status. An LLM judge validated against human annotators scores each turn on five teaching metrics. Paired statistical tests measure bias.

AgentFairBench is a multi-domain benchmark for measuring demographic disparity in the actions produced by LLM agents, rather than only the fairness of their written answers. It evaluates whether agent...

AI Bias Auditing Tool is a Python tool for detecting bias in tabular datasets. Users upload a CSV file and choose a sensitive attribute and an outcome column. The tool calculates approval rates for each group. It then measures the Demographic Parity Difference using Fairlearn. Results are shown as charts and a short bias audit report. The tool runs in Google Colab and is designed for beginners.

The VeritasChain Protocol (VCP) is an open specification for cryptographically verifiable audit trails in algorithmic and AI-driven trading. It is maintained by the VeritasChain Standards Organization. VCP lets regulators and auditors verify mathematically that records of trading decisions, orders, risk controls and AI model use are complete and unaltered. A privacy module allows personal data to be erased while keeping the audit trail intact. VCP is designed to support MiFID II, EU AI Act and GDPR compliance.

AiEnsured is a commercial testing platform for AI products, designed to support the deployment of responsible AI. It is developed in India. The platform brings together fairness and bias evaluation, explainability, adversarial robustness testing and privacy and GDPR compliance checks. It also generates test cases automatically, including rare and extreme inputs. Further features cover experiment management, model comparison and performance testing. After deployment, it monitors models for concept drift, which can reduce accuracy over time.

F2Bench is a benchmark for evaluating fairness in large language models that also considers factual accuracy. It was developed by researchers at Inner Mongolia University and Xiamen University. The benchmark contains 2 568 test instances across ten demographic categories, including intersectional combinations. One task tests whether models accept biased conclusions after gradual stereotypical suggestions in a conversation. A second task tests whether models avoid stereotypes while still reflecting real statistics accurately. The code and dataset are publicly available.

MetricLib is an open-source Python toolkit for evaluating the quality of medical machine learning datasets. It was developed by the Physikalisch-Technische Bundesanstalt (PTB) and is based on the METRIC-framework for trustworthy AI in medicine. The toolkit implements data quality metrics across five clusters, from measurement process to consistency. Users generate reports that help identify data quality gaps. They can also add custom metrics for their own use cases. The code is available under the GPL-2.0 licence.

Metric Hub is an online platform developed by PTB and partners for assessing the quality of data used in medical AI. Its metric library of 60 data quality metrics puts into practice the METRIC-framework, and each metric comes with a standardised card describing its definition, applicability and pitfalls. Decision trees guide users to the metrics that suit their use case. This enables systematic, fit-for-purpose evaluation of training and test data, supporting trustworthy AI and regulatory compliance.

EuConform is an open-source toolkit and open specification for producing technical evidence of compliance with the EU AI Act. It scans an AI system's code and produces machine-readable evidence files, including an AI bill of materials and a report of compliance gaps. Files are bundled with cryptographic hashes so they can be verified. It also offers AI Act risk classification and offline bias testing of language models. It provides technical guidance only, not legal advice.

PrivFair is a library for auditing the fairness of machine learning models while protecting privacy. Fairness audits require sensitive data about attributes such as gender or race, which is protected by law. PrivFair uses Secure Multiparty Computation, a cryptographic technique, so that an external auditor can test a company's model without either party revealing their data or model. It supports group fairness audits on tabular and image data. This enables independent, privacy-preserving checks of proprietary AI models.

DP-CGANs is an open-source Python library for generating synthetic data that protects individual privacy, designed with personal health data in mind. It was developed at Maastricht University. The library uses a conditional generative adversarial network, adapted to capture relationships between variables in the data. Differential privacy can be enabled during training, so no single person strongly influences the result. It works with tabular and RDF data. It is still under development and released under the MIT licence.

MANILA is a web-based, low-code application for benchmarking machine learning models and methods that reduce bias. It helps users find the combination that offers the best balance between fairness and performance. MANILA models a fairness benchmarking workflow as a set of options, such as datasets, models and metrics. Rules between these options guide users and prevent experiments that would fail. This makes fairness benchmarking accessible to users with little or no programming experience.

Fairmetrics is an open-source R package for evaluating the group fairness of machine learning models. It was developed by researchers at the University of Toronto. The package calculates a wide range of fairness metrics, such as statistical parity and equal opportunity. Unlike most fairness tools, it also provides confidence intervals for each metric. This shows whether differences between groups are statistically significant or due to random variation. It is available on CRAN under the MIT licence.

IEEE 3198-2025 is a technical standard that specifies a method for evaluating the fairness of machine learning systems. It was published by the IEEE Standards Association in May 2025. The standard categorises the causes of unfairness in machine learning and presents widely used definitions of fairness. For each definition, it specifies metrics and explains how to calculate them. Test cases set out detailed procedures for fairness evaluations. This supports consistent, repeatable fairness testing across systems and organisations.

FairX is an open-source toolkit for benchmarking machine learning models on fairness, data utility and explainability. It was developed by researchers at Linköping University. Users load tabular or image datasets and apply bias-mitigation methods before, during or after model training, including fair generative models that create synthetic data. FairX evaluates results using fairness, performance and synthetic data quality metrics. This lets users compare trade-offs between fairness and utility. It is written in Python and released under the MIT licence.

Google's Differential Privacy libraries are an open-source collection of tools for producing statistics from sensitive data while protecting individuals' privacy. They add calibrated random noise to results such as counts and averages, so no single person can be identified. Building block libraries are available in C++, Go and Java. End-to-end frameworks make the tools usable by non-experts on large-scale data systems. Further tools track privacy budgets and audit privacy guarantees. The libraries are released under the Apache 2.0 licence.

Audit the Judge is an open-source tool for checking whether large language models used as judges evaluate answers reliably. It builds answer pairs where the correct verdict is known. These test whether the judge invents preferences between equal answers or favours answers shown first or written at greater length. Results are presented in a one-page report with statistical confidence intervals. The tool works with any major LLM judge. It is written in Python and released under the MIT licence.