Awesome Responsible AI

May 28, 2026 · View on GitHub

Awesome Maintenance GitHub GitHub GitHub GitHub

Awesome Responsible AI

A curated list of awesome academic research, books, code of ethics, courses, data sets, databases, frameworks, institutes, maturity models, newsletters, principles, podcasts, regulations, responsible scale policies, reports, tools and standards related to Responsible, Trustworthy, and Human-Centered AI.

Main Concepts

What is AI Governance?

AI governance is a system of rules, processes, frameworks, and tools within an organization to ensure the ethical and responsible development of AI.

What is Human-Centered AI?

Human-Centered Artificial Intelligence (HCAI) is an approach to AI development that prioritizes human users' needs, experiences, and well-being.

What is Open Source AI?

When we refer to a “system,” we are speaking both broadly about a fully functional structure and its discrete structural elements. To be considered Open Source, the requirements are the same, whether applied to a system, a model, weights and parameters, or other structural elements.

An Open Source AI is an AI system made available under terms and in a way that grant the freedoms1 to:

  • Use the system for any purpose and without having to ask for permission.
  • Study how the system works and inspect its components.
  • Modify the system for any purpose, including to change its output.
  • Share the system for others to use with or without modifications, for any purpose.

Source

What is Responsible AI?

Responsible AI (RAI) refers to the development, deployment, and use of artificial intelligence (AI) systems in ways that are ethical, transparent, accountable, and aligned with human values.

What is a Responsible AI framework?

Responsible AI frameworks often encompass guidelines, principles, and practices that prioritize fairness, safety, and respect for individual rights.

What is Trustworthy AI?

Trustworthy AI (TAI) refers to artificial intelligence systems designed and deployed to be transparent, robust and respectful of data privacy.

Why is Responsible, Trustworthy, and Human-Centered AI important?

AI is a transformative and dual-side technology prone to reshape industries, yet it requires careful governance to balance the benefits of automation and insight with protections against unintended social, economic, and security impacts. You can read more about the current wave here.

Content

Academic Research

Adversarial ML

  • Oprea, A. et al. (2023). Adversarial machine learning: A taxonomy and terminology of attacks and mitigations. National Institute of Standards and Technology. Article

Artificial General Intelligence (AGI)

  • Hendricks, D. et al. (2025). A definition of AGI. Article

Artificial Intelligence Governance (AI Governance)

  • Eisenberg, I. W. et al. (2025). The Unified Control Framework: Establishing a Common Foundation for Enterprise AI Governance, Risk Management and Regulatory Compliance. arXiv preprint arXiv:2503.05937. Article Visualization Credo

Bias

  • Schwartz, R. et al. (2022). Towards a standard for identifying and managing bias in artificial intelligence (Vol. 3, p. 00). US Department of Commerce, National Institute of Standards and Technology. Article NIST

Challenges

  • D'Amour, A. et al. (2022). Underspecification presents challenges for credibility in modern machine learning. Journal of Machine Learning Research, 23(226), 1-61. Article Google

Drift

  • Ackerman, S. et al. (2021, June). Machine learning model drift detection via weak data slices. In 2021 IEEE/ACM Third International Workshop on Deep Learning for Testing and Testing for Deep Learning (DeepTest) (pp. 1-8). IEEE. Article IBM
  • Ackerman, S. et al. (2020, February). FreaAI: Automated extraction of data slices to test machine learning models. In International Workshop on Engineering Dependable and Secure Machine Learning Systems (pp. 67-83). Cham: Springer International Publishing. Article IBM

Explainability/Interpretability/Mechanical Interpretability

  • Dhurandhar, A. et al. (2018). Explanations based on the missing: Towards contrastive explanations with pertinent negatives. Advances in neural information processing systems, 31. Article University of Michigan IBM Research
  • Dhurandhar, A. et al. (2018). Improving simple models with confidence profiles. Advances in Neural Information Processing Systems, 31. Article IBM Research
  • Gurumoorthy, K. S. et al. (2019, November). Efficient data representation by selecting prototypes with importance weights. In 2019 IEEE International Conference on Data Mining (ICDM) (pp. 260-269). IEEE. Article Amazon Development Center IBM Research
  • Hind, M. et al. (2019, January). TED: Teaching AI to explain its decisions. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society (pp. 123-129)Article IBM Research
  • Lundberg, S. M. et al. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, 30. Article, Github University of Washington
  • Luss, R. et al. (2021, August). Leveraging latent features for local explanations. In Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining (pp. 1139-1149). Article IBM Research University of Michigan
  • Ribeiro, M. T. et al. (2016, August). "Why should i trust you?" Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1135-1144). Article, Github University of Washington
  • Wei, D. et al. (2019, May). Generalized linear rule models. In International conference on machine learning (pp. 6687-6696). PMLR. Article IBM Research
  • Contrastive Explanations Method with Monotonic Attribute Functions (Luss et al., 2019)
  • Boolean Decision Rules via Column Generation (Light Edition) (Dash et al., 2018) IBM Research
  • Towards Robust Interpretability with Self-Explaining Neural Networks (Alvarez-Melis et al., 2018) MIT

An interesting curated collection of articules (updated until 2021) A Living and Curated Collection of Explainable AI Methods.

A shared effort can be found at Neuronpedia.

Ethical Data Products

  • Gebru, T. et al. (2021). Datasheets for datasets. Communications of the ACM, 64(12), 86-92. Article Google
  • Mitchell, M. et al. (2019, January). Model cards for model reporting. In Proceedings of the conference on fairness, accountability, and transparency (pp. 220-229). Article Google
  • Pushkarna, M. et al. (2022, June). Data cards: Purposeful and transparent dataset documentation for responsible ai. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (pp. 1776-1826). Article Google
  • Rostamzadeh, N. et al. (2022, June). Healthsheet: development of a transparency artifact for health datasets. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (pp. 1943-1961). Article Google
  • Saint-Jacques, G. et al. (2020). Fairness through Experimentation: Inequality in A/B testing as an approach to responsible design. arXiv preprint arXiv:2002.05819. Article LinkedIn

Evaluation (of model explanations)

  • Agarwal, C. et al. (2022). Openxai: Towards a transparent evaluation of model explanations. Advances in Neural Information Processing Systems, 35, 15784-15799. Article
  • Liesenfeld, A. et al. (2024). Rethinking Open Source Generative AI: Open-Washing and the EU AI Act. In The 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’24). Rio de Janeiro, Brazil: ACM. Article Benchmark

Fairness

  • Caton, S. et al. (2024). Fairness in machine learning: A survey. ACM Computing Surveys, 56(7), 1-38. Article
  • Chouldechova, A. (2017). Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big data, 5(2), 153-163. Article
  • Chouldechova, A. et al. (2017). Fairer and more accurate, but for whom? arXiv preprint arXiv:1707.00046. Article
  • Coston, A. et al. (2020, January). Counterfactual risk assessments, evaluation, and fairness. In Proceedings of the 2020 conference on fairness, accountability, and transparency (pp. 582-593). Article
  • Kleinberg, J. et al. (2016). Inherent trade-offs in the fair determination of risk scores. arXiv preprint arXiv:1609.05807. Article
  • Saleiro, P. et al. (2024). Aequitas Flow: Streamlining Fair ML Experimentation. arXiv preprint arXiv:2405.05809. Article
  • Plečko, D. et al. (2024), Causal Fairness Analysis, Foundations and Trends in Machine Learning: Vol. 17, No. 3, pp 1–238. DOI: 10.1561/2200000106. Article and Materials
  • Saleiro, P. et al. (2018). Aequitas: A bias and fairness audit toolkit. arXiv preprint arXiv:1811.05577. Article
  • Vasudevan, S. et al. (2020, October). Lift: A scalable framework for measuring fairness in ml applications. In Proceedings of the 29th ACM international conference on information & knowledge management (pp. 2773-2780). Article LinkedIn

Regulation

  • Wasil, A. R. et al. (2024). Verification methods for international AI agreements. arXiv preprint arXiv:2408.16074. Article

Representation Engineering

  • Zou, A. et al. (2024) Improving Alignment and Robustness with Circuit Breakers. Article
  • Zou, A. et al. (2023) Representation Engineering: A Top-Down Approach to AI Transparency. Article

Risk

  • Slattery, P., et al. (2024). The ai risk repository: A comprehensive meta-review, database, and taxonomy of risks from artificial intelligence. arXiv preprint arXiv:2408.12622. Article

Systems Risks

  • Uuk, R., et al. (2024). A Taxonomy of Systemic Risks from General-Purpose AI. arXiv preprint arXiv:2412.07780. Article

Sustainability

  • Lacoste, A. et al. (2019). Quantifying the carbon emissions of machine learning. arXiv preprint arXiv:1910.09700. Article
  • Li, P. et al. (2023) Making AI Less “Thirsty”: Uncovering and Addressing the Secret Water Footprint of AI Models. arXiv:2304.03271 Article
  • Parcollet, T. et al. (2021). The energy and carbon footprint of training end-to-end speech recognizers. Article
  • Patterson, D. et al. (2021). Carbon emissions and large neural network training. arXiv preprint arXiv:2104.10350. Article
  • Sculley, D. et al. (2015). Hidden technical debt in machine learning systems. Advances in neural information processing systems, 28. Article Google
  • Sculley, D. et al. (2014, December). Machine learning: The high interest credit card of technical debt. In SE4ML: software engineering for machine learning (NIPS 2014 Workshop) (Vol. 111, p. 112). Article Google
  • Strubell, E. et al. (2019). Energy and policy considerations for deep learning in NLP. arXiv preprint arXiv:1906.02243. Article
  • Sustainable AI: AI for sustainability and the sustainability of AI (van Wynsberghe, A. 2021). AI and Ethics, 1-6
  • Green Algorithms: Quantifying the carbon emissions of computation (Lannelongue, L. et al. 2020)
  • C.-J. Wu, R. Raghavendra, U. Gupta, B. Acun, N. Ardalani, K. Maeng, G. Chang, F. Aga, J. Huang, C. Bai, M. Gschwind, A. Gupta, M. Ott, A. Melnikov, S. Candido, D. Brooks, G. Chauhan, B. Lee, H.-H. Lee, K. Hazelwood, Sustainable AI: Environmental implications, challenges and opportunities. Proceedings of the 5th Conference on Machine Learning and Systems (MLSys) (2022) vol. 4, pp. 795–813. Article

Collections

Reproducible/Non-Reproducible Research

Computational reproducibility (when the results in a paper can be replicated using the exact code and dataset provided by the authors) is becoming a significant problem not only for academic but for practitionars who want to implement AI in their organizations and aim to resuse ideas from academia. Read more about this problem here.

Books

This section features a curated selection of books.

Open Access

  • Barocas, S., Hardt, M., & Narayanan, A. (2023). Fairness and machine learning: Limitations and opportunities. MIT press. Book
  • Barrett, M., Gerke, T. & D’Agostino McGowa, L. (2024). Causal Inference in R Book Causal Inference R
  • Biecek, P., & Burzykowski, T. (2021). Explanatory model analysis: explore, explain, and examine predictive models. Chapman and Hall/CRC. Book Explainability Interpretability Transparency R
  • Biecek, P. (2024). Adversarial Model Analysis. Book Safety Red Teaming
  • Cunningham, Scott. (2021) Causal inference: The mixtape. Yale university press. Book Causal Inference
  • Fourrier, C. and et all. (2024) LLM Evaluation Guidebook. Github Repository. Web LLM Evaluation
  • Freiesleben, T. & Molnar, C. (2024). Supervised Machine Learning for Science: How to stop worrying and love your black box. Book
  • Huntington-Klein, N. (2012) The effect: An introduction to research design and causality. Chapman and Hall/CRC. Book Causal Inference
  • Leveson, N. G. (2016). Engineering a safer world: Systems thinking applied to safety. MIT press. Book
  • Kamath, U. et al. (2023) Applied Causal Inference Book
  • Matloff, N et al. (2204) Data Science Looks at Discrimination Book Fairness R
  • Molnar, C. (2020). Interpretable Machine Learning. Lulu.com. Book Explainability Interpretability Transparency R
  • Tegomoh, B. (2025). The Biosecurity Handbook: Biological Security in the AI Era. Book
  • Tegomoh, B. (2025). The Public Health AI Handbook: Evaluating AI Tools for Public Health Practice. Book
  • Vizquez, S. & Kubersky, W. (2025) The Little Book of ML Metrics. Book ML Evaluation

Commercial / Propietary / Closed Access

Code of Ethics

This section features a curated selection of code of ethics.

Courses

This section features a curated selection of open courses focused on Responsible AI, AI Ethics, AI Safety and other related topics. The classes range from introductory courses on data ethics to specialized training in AI Safety.

CourseOrganizationDescriptionTopic
AGI StrategyBlueDot ImpactA course abour AGI to understand to understand the race, the risks, and how you can make a difference.AGI Strategy
AI AlignmentBlueDot ImpactA course designed to introduce the key concepts in AI safety and alignment.AI Alignment, AI Safety
AI EthicsTuring CollegeThis course is part of the DIVERSIFAIR project, an EU-backed initiative created to help professionals build ethical AI that’s fair, transparent, and accountable — not just technically accurate.AI Ethics, Fairness
AI Ethics & Governance (AEG)The Alan Turing InstituteThis course is designed to help you understand the fundamentals of AI Ethics and Governance.AI Ethics, AI Governance
AI GovernanceAI Career ProA series of courses that teach the practical capabilities missing from AI governance education — not just theory, but how to actually build AI inventories, perform algorithmic assessments, design meaningful human oversight, and make the business case that secures resources.AI Governance
AI GovernanceBlueDot ImpactA course designed to Examine the risks posed by advanced AI systems, standards and regulations to address them, and foreign policy approaches.AI Governance
AI Policy ClinicCenter for AI and Digital PolicyThe Center has launched a comprehensive certification program for AI Policy.AI Governance
AI Safety, Ethics and SocietyCenter for AI SafetyA course aims to provide a comprehensive introduction to how current AI systems work, why many experts are concerned that continued advances in AI may pose severe societal-scale risks, and how society can manage and mitigate these risks.AI Safety, AI Ethics, AI Governance
AI Security and GovernanceSecuritiThis certification covers core concepts in generative AI, global AI laws, compliance obligations, AI risk management, and AI governance frameworks that ensure responsible innovation.AI Security, AI Governance
CS 2881 AI SafetyHarvard UniversityThis course introduces challenges in alignment and safety of artificial intelligence.AI Safety
CS 294-131: Trustworthy Deep LearningBerkeley UniversityThis course helps to develop a deeper understanding of deep learning and explore new research directions and applications of AI/deep learning and privacy/security.Explainability, Privacy, Security
CIS 4230/5230 - Ethical Algorithm DesignUniversity of PennsylvaniaThis course is about the social and human problems that can arise from algorithms, AI and machine learning, and how we might design these technologies to be "better behaved" in the first place.AI Safety, Responsible AI
CS 594 - Causal Inference and LearningUniversity of Illinois at ChicagoThe goal of the course on Causal is to introduce students to methodologies and algorithms for causal reasoning and connect various aspects of causal inference, including methods developed within computer science, statistics, and economics.Causal Inference
CS 7880 - Rigorous Approaches to Data PrivacyNortheastern UniversityThis course covers the theory of differential privacy, its application, and its connections to other areas of computer science, covering roughly the state-of-the-art in the field.Data Privacy
CS 860 - Algorithms for Private Data AnalysisUniversity of WaterlooThis course is on algorithms for differentially private analysis of data.Data Privacy
Data Justice (DJ)The Alan Turing InstituteA course that explores the emerging movement of data justice, which seeks to apply a social justice-oriented approach to examining the range of social, political, and material concerns arising within our increasingly datafied societyEthics, Data Justice
Explainable Artificial IntelligenceHarvard UniversityThis course aims to familiarize students with the recent advances in the emerging field of eXplainable Artificial Intelligence (XAI)Explainability, Interpretability
Future of AIBlueDot ImpactA course to understand AI's impact and be part of the conversation about its future.AI Fundamentals
Introduction to AI EthicsKaggleA course to explore practical tools to guide the moral design of AI systems.AI Ethics
Introduction to ML SafetyCenter for AI SafetyA course discusses how researchers can shape the process that will lead to strong AI systems and steer that process in a safer direction.AI Safety
Introduction to Responsible Machine LearningThe George Washington UniversityMaterials for a technical, nuts-and-bolts course about increasing transparency, fairness, robustness, and security in machine learning.Responsible AI
LLM evaluationNebius Academy, EvidentlyA course about LLM evaluation using Evidently.AI Safety, LLM Evaluation
Machine Learning ExplainabilityKaggleA course to extract human-understandable insights from any model.Explainability, Interpretability
Machine Learning in Production (17-445/17-645/17-745) / AI Engineering (11-695)Carnegie Mellon UniversityA course that covers how to build, deploy, assure, and maintain software products with machine-learned models.MLOps, Responsible AI
MATSMATS ResearchThe main goal of the course is to help scholars develop as AI alignment researchers.AI Alignment, AI Safety
Modern-Day Oracles or Bullshit Machines?Bergstrom, C. T., & West, J. D.A course about how data and statistical analysis — the keystones of scientific reasoning — can be abused to mislead people.Ethics
Practical Data EthicsFast.aiA course focus on topics that are both urgent and practical.Data Ethics
Public Engagement of Data Science and AI (PED)The Alan Turing InstituteA course is designed to help you understand the practical and ethical value of public engagement with data science and AI.Ethics
Responsible AIAll Tech is HumanThis series of short courses, which can be completed in just a few hours, offers a foundational understanding of Responsible AI.Responsible AI
Responsible Research and Innovation (RRI)The Alan Turing InstituteThis course explores what it means to take (individual and collective) responsibility for (and over) the processes and outcomes of research and innovation in data science and AI.Responsible AI

Data Sets

This section features a curated selection of data sets.

If you are looking for public data sets for your project, this is a curated collection.

Databases

This section features a curated selection of databases focused on tracking incidents, issues, litigations, vulnerabilities and Ai for good initiatives.

(AI) Incidents Trackers

TrackerPaperOrganization/CreatorDescriptionTopic
AI for Good LabN/AMicrosoftAn open source database of assets for social and environmental good.AI for Good
AI Hallucination CasesN/ADamien CharlotinThis database tracks legal decisions1 in cases where generative AI produced hallucinated content – typically fake citations, but also other types of arguments.Deepfakes, Misinformation
AI Risk RepositoryThe AI Risk Repository: A Comprehensive Meta-Review, Database, and Taxonomy of Risks From Artificial IntelligenceMITA comprehensive living database of over 1600 AI risks categorized by their cause and risk domain.AI Risk
Political Deepfakes Incidents DatabaseMerging AI Incidents Research with Political Misinformation Research: Introducing the Political Deepfakes Incidents DatabasePurdue UniversityA collection of politically-salient deepfakes, encompassing synthetically-created videos, images, and less-sophisticated `cheapfakes.'Deepfakes

This section is under review and the rest of entries will be added to the table with extended information.

Cybersecurity

Frameworks

Institutes

This section features a curated selection of institutes that research about Responsible AI and related topics.

AI Safety Institutes (or equivalent)

AI Security Institute

Japan AISI

CodeTitleDescriptionStatusSource
AI Safety Evaluation v1.10A guide to red teaming techniques for AI safetyPresents basic concepts that those involved in the development and provision of AI systems can refer to when conducting AI Safety evaluationsPublishedSource
AI Safety RT v1.10Guide to Red Teaming Methodology on AI SafetyIntended to help developers and providers of AI systems to evaluate the basic considerations of red teaming methodologies for AI systems from the viewpoint of attackersPublishedSource
Data Quality Management v1.0.0A guide about Data Quality linked to AI SafetyIntended to help developers and providers of AI systems to adopt data quality management practicesPublishedSource
AI Business Guidelines v1.1.0A guide for organizations to adopt agile AI GovernanceIntended to help all the stakeholders in an organization to adopt voluntary agile AI Governance practicesPublishedSource
Known Attacks and Their Impacts on AI Systems (March 2025)Known Attacks and Their Impacts on AI SystemsAn accessible overview of adversarial attacks unique to predictive and generative AI systemsPublishedSource 1, Source 2

US CAISI

CodeTitleDescriptionStatusSource
NIST AI 800-1Managing Misuse Risk for Dual-Use Foundation ModelsOutlines voluntary best practices for identifying, measuring, and mitigating risks to public safety and national security across the AI lifecycleDraft (second Version)Source

Responsible AI Institutes

Research Institutes

Maturity Models

This section features a curated selection of maturity models that can help organizations to adopt AI in a responsible way.

AI Governance

Ethics

Responsible AI

Newsletters

This section features a curated selection of newsletters that keep you informed about this domain.

Principles

This section features a curated selection of Responsible AI principles adopted by several organizations. Note: some publication dates may not be accurate.

CompanyDocumentYear
AllianzPrinciples for a responsible usage of AI2022
Deutsche TelekomGuidelines for Artificial Intelligence2018
European CommissionGuidelines for Trustworthy AI2019
Future of Life InstituteAsilomar AI principles2017
GoogleAI Principles2019
IEEEEthically Aligned Design2023
LogitechPrinciples for Responsible AI2024
MicrosoftAI Principles2022
OECDAI principles2019
TelefonicaAI Principles2018
The Institute for Ethical AI & Machine LearningThe Responsible Machine Learning Principles2018

Additional:

If you want to read on how to move from AI principles to commitments, we recommend hte article 'Getting from commitment to content in AI and data ethics: Justice and explainability'.

Podcasts

This section features podcasts that offer insightful commentary and explanations on Responsible AI, AI Governance, AI Safety, AI Alignment and Machine Learning Interpretability.

PodcastDescriptionCreator
AI FrontiersA space for expert dialogue about the impact of AI.Center for AI Safety
AI Safety FundamentalsListen to the Bluedot Impact courses contentBlueDot Impact
AI Safety NewsletterNarrations of the newsletter.Center for AI Safety
Me, Myself and AIInterviews with expertsMIT Sloan Management Review
Practical AIPractical AI is a show in which technology professionals, business people, students, enthusiasts, and expert guests engage in lively discussions about Artificial Intelligence and related topics.Chris Benson

Regulations

This section features a curated selection of regulations.

Definition

What are regulations?

Regulations are requirements established by governments.

Interesting resources

Australia 🇦🇺

Additionally, we recommend this comprehensive, community-maintained index of Australian AI Security standards, policies, frameworks, and guidance.

Canada 🇨🇦

China 🇨🇳

European Union 🇪🇺

Short NameCodeDescriptionStatusWebsiteLegal text
Cyber Resilience Act (CRA) - horizontal cybersecurity requirements for products with digital elements2022/0272(COD)It introduces mandatory cybersecurity requirements for hardware and software products, throughout their whole lifecycle.ProposalWebsiteSource
Data ActEU/2023/2854It enables a fair distribution of the value of data by establishing clear and fair rules for accessing and using data within the European data economy.PublishedWebsiteSource
Data Governance ActEU/2022/868It supports the setup and development of Common European Data Spaces in strategic domains, involving both private and public players, in sectors such as health, environment, energy, agriculture, mobility, finance, manufacturing, public administration and skills.PublishedWebsiteSource
Digital Market ActEU/2022/1925It establishes a set of clearly defined objective criteria to identify “gatekeepers”. Gatekeepers are large digital platforms providing so called core platform services, such as for example online search engines, app stores, messenger services. Gatekeepers will have to comply with the do’s (i.e. obligations) and don’ts (i.e. prohibitions) listed in the DMA.PublishedWebsiteSource
Digital Operational Resilience Act (DORA)EU/2022/2554IT is a regulation to strengthen the digital resilience of financial entities, and ensures that banks, insurance companies, investment firms and other financial entities can withstand, respond to, and recover from ICT (Information and Communication Technology) disruptions, such as cyberattacks or system failures.PublishedWebsiteSource
Digital Services ActEU/2022/2026It regulates online intermediaries and platforms such as marketplaces, social networks, content-sharing platforms, app stores, and online travel and accommodation platforms. Its main goal is to prevent illegal and harmful activities online and the spread of disinformation. It ensures user safety, protects fundamental rights, and creates a fair and open online platform environment.PublishedWebsiteSource
DMS DirectiveEU/2019/790It is intended to ensure a well-functioning marketplace for copyright.PublishedWebsiteSource
Energy Efficiency DirectiveEU/2023/1791It establishes ‘energy efficiency first’ as a fundamental principle of EU energy policy, giving it legal-standing for the first time. In practical terms, this means that energy efficiency must be considered by EU countries in all relevant policy and major investment decisions taken in the energy and non-energy sectors.PublishedWebsiteSource
EU AI ACTEU/2024/1689It assigns applications of AI to three risk categories. First, applications and systems that create an unacceptable risk are banned. Second, high-risk applications are subject to specific legal requirements. Lastly, applications not explicitly banned or listed as high-risk are largely left unregulated.PublishedWebsiteSource
General Data Protection Regulation (GDPR)EU/2016/679It strengthens individuals' fundamental rights in the digital age and facilitate business by clarifying rules for companies and public bodies in the digital single market.PublishedWebsiteSource
NIS2 DirectiveEU/2022/2555It provides legal measures to boost the overall level of cybersecurity in the EU by ensuring preparedness, cooperation and security cultere across the Member States.PublishedWebsiteSource

Additionally,

India 🇮🇳

Singapore 🇸🇬

South Korea 🇰🇷

United Arab Emirates 🇦🇪

United States 🇺🇸

Spain 🇪🇸

Responsible Scaling Policies

This section features a curated selection of responsible scale policies adopted by AI Labs and organizations developing frontier models.

Definition

Responsible Scale Policies (RSPs) specify what level of AI capabilities an AI developer is prepared to handle safely with their current protective measures, and conditions under which it would be too dangerous to continue deploying AI systems and/or scaling up AI capabilities until protective measures improve.

RSP List

Reports

This section features a curated selection of reports relevant to understand the current situation and trends related to Responsible AI, AI Ethics and AI Governance.

AI Ethics

  • State of AI Ethics. MAIEI Website (latest version 7, November 2025)

AI Governance

  • Araujo, R. 2024. Understanding the First Wave of AI Safety Institutes: Characteristics, Functions, and Challenges. Institute for AI Policy and Strategy (IAPS) Article
  • Buchanan, B. 2020. The AI triad and what it means for national security strategy. Center for Security and Emerging Technology. Article
  • Corrigan, J. et al. 2023. The Policy Playbook: Building a Systems-Oriented Approach to Technology and National Security Policy. CSET (Center for Security and Emerging Technology) Article
  • Curto, J. 2024. How Can Spain Remain Internationally Competitive in AI under EU Legislation? Article
  • CSIS. 2024 The AI Safety Institute International Network: Next Steps and Recommendations. CSIS (Center for Strategic and International Studies) Article
  • Gupta, Ritwik, et al. (2024). Data-Centric AI Governance: Addressing the Limitations of Model-Focused Policies. arXiv preprint arXiv:2409.17216 (Article)[https://arxiv.org/pdf/2409.17216]
  • Hendrycks, D. et al. 2023. An overview of catastrophic AI risks. Center for AI Safety. arXiv preprint arXiv:2306.12001. Article
  • Janjeva, A., et al. (2023). Strengthening Resilience to AI Risk. A guide for UK policymakers. CETaS (Centre for Emerging Technology and Security) Article
  • Piattini, M. and Fernández C.M. 2024. Marco Confiable. Revista SIC 162 Article
  • Sastry, G., et al. 2024. Computing Power and the Governance of Artificial Intelligence. arXiv preprint arXiv:2402.08797. Article

AI Safety

AI Security

AI Testing

Market Analysis

AI Labs

Other

Ratings

Standards

Definition

What are standards?

Standards are voluntary, consensus solutions. They document an agreement on how a material, product, process, or service should be specified, performed or delivered. They keep people safe and ensure things work. They create confidence and provide security for investment.

Standards can be understood as formal specifications of best practices as well. There is a growing number of standards related to AI. You can search for the latest in the Standards Database from AI Standards Hub.

There are some open standards such as RSL, focused on content licensing, that still need to gain traction in the market.

Standards

This section features a curated selection of standards related to Responsible AI.

CEN Standards

The European Committee for Standardization is one of three European Standardization Organizations (together with CENELEC and ETSI) that have been officially recognized by the European Union and by the European Free Trade Association (EFTA) as being responsible for developing and defining voluntary standards at European level.

DomainStandardStatusURL
Data governance and quality for AI within the European contextCEN/CLC/TR 18115:2024PublishedSource

CEN AI Work programme can be found here.

DGSI Standards

The Digital Governance Standards Institute, part of the Digital Governance Council, is an accredited standards development body. The Institute enables greater trust and confidence in Canada’s digital systems through developing technology governance standards collaboratively across a range of stakeholders.

DomainStandardStatusURL
Design, Implementation and Evaluation of Regulatory SandboxesCAN/DGSI 123PublishedSource
Ethical Design and Use of Artificial Intelligence by Small and Medium OrganizationsCAN/DGSI 101PublishedSource
Machine Learning and AI Implementation in Research InstitutionsCAN/DGSI 128PublishedSource

ETSI Standards

The European Telecommunications Standards Institute (ETSI) is an independent, not-for-profit, standardization organization operating in the field of information and communications.

DomainStandardStatusURL
Securing Artificial Intelligence (SAI); Baseline Cyber Security Requirements for AI Models and SystemsETSI EN 304 223 V2.1.1 (2025-12)PublishedSource
Securing Artificial Intelligence (SAI); AI Threat OntologyETSI GR SAI 001 V1.1.1 (2022-01)PublishedSource
Securing Artificial Intelligence (SAI); Explicability and transparency of AI processingETSI GR SAI 007 V1.1.1 (2023-03)PublishedSource
Securing Artificial Intelligence (SAI); Artificial Intelligence Computing Platform Security FrameworkETSI GR SAI 009 V1.1.1 (2023-02)PublishedSource
Securing ArtificiaI Intelligence (SAI);The role of hardware in security of AIETSI GR SAI 006 V1.1.1 (2022-03)PublishedSource
Securing Artificial Intelligence TC (SAI);Privacy aspects of AI/ML systemsETSI TR 104 225 V1.1.1 (2024-04)PublishedSource
Securing Artificial Intelligence (SAI); Explicability and transparency of AI processingETSI TS 104 224 V1.1.1 (2025-03)PublishedSource
Methods for Testing & Specification (MTS); AI Testing; Guidelines for Documentation of AI-enabled SystemsETSI TR 104 119 V1.1.1 (2025-09)PublishedSource
Securing Artificial Intelligence;Security Testing of AIETSI TR 104 066 V1.1.1 (2024-07)PublishedSource
Securing Artificial Intelligence (SAI);Baseline Cyber Security Requirements for AI Models and SystemsETSI TS 104 223 V1.1.1 (2025-04)PublishedSource
Securing Artificial Intelligence (SAI);Baseline Cyber Security Requirements for AI Models and SystemsETSI EN 304 223 V2.1.1 (2025-12)PublishedSource
Securing Artificial Intelligence (SAI); AI Threat Ontology and definitionsETSI TS 104 050 V1.1.1 (2025-03)PublishedSource
Securing Artificial Intelligence (SAI);Traceability of AI ModelsETSI TR 104 032 V1.1.1 (2024-02)PublishedSource
Methods for Testing & Specification (MTS);Continuous Auditing Based Conformity Assessment for AI-enabled systemsETSI TS 104 008 V1.1.1 (2026-01)PublishedSource
Securing Artificial Intelligence (SAI); Guide to Cyber Security for AI Models and SystemsETSI TR 104 128 V1.1.1 (2025-05)PublishedSource
Securing Artificial Intelligence (SAI); Understanding and Preventing Harm from Generative AIETSI TR 104 159 V1.1.1 (2026-01)PublishedSource
Securing Artificial Intelligence (SAI); Proofs of Concepts FrameworkETSI GR SAI 013 V1.1.1 (2023-03)PublishedSource

IEEE Standards

The Institute of Electrical and Electronics Engineers (IEEE) is an American 501 charitable professional organization for electrical engineering, electronics engineering, and related disciplines. Today, it is a global network of more than 486,000 engineering and STEM professionals across a variety of disciplines whose core purpose is to foster technological innovation for the benefit of humanity.

DomainStandardStatusURL
IEEE Guide for an Architectural Framework for Explainable Artificial IntelligenceIEEE 2894-2024PublishedSource
IEEE Recommended Practice for the Quality Management of Datasets for Medical Artificial IntelligenceIEEE 2801-2022PublishedSource
IEEE Standard for Ethical Considerations in Emulated Empathy in Autonomous and Intelligent SystemsIEEE 7014-2024PublishedSource
IEEE Standard for Robustness Testing and Evaluation of Artificial Intelligence (AI)-based Image Recognition ServiceIEEE 3129-2023PublishedSource
IEEE Standard for Performance Benchmarking for Artificial Intelligence Server SystemsIEEE 2937-2022PublishedSource
IEEE Standard for Security Requirement of Privacy-Preserving ComputationIEEE 3169-2025PublishedSource

SAE Standards

SAE International is community of more than 200,000 aerospace, commercial vehicle, and ground vehicle engineers and technical experts, SAE solves challenges, creates connections, and develops resources that allow us to drive industry forward.

DomainStandardStatusURL
Artificial intelligence Simulation: Best Practices3347WIPSource
Assessment of Human Factors concerns for the development of safety-related AI-based Systems in AviationAIR8493WIPSource
Verification & Validation of AI/ML based Components & Systems in Ground VehiclesJ3321WIPSource
AI Regulations, Standards & Applications ChallengesJ3329WIPSource
Managing the Development of Artificial Intelligence SoftwareCRB1Publishedsource
Artificial Intelligence in Aeronautical Systems: Statement of ConcernsAIR6988PublishedSource

UNE Standards

UNE is Spain's only Standardisation Organisation, designated by the Spanish Ministry of Economy, Industry and Competitiveness to the European Commission. It helps Spanish organizations to keep up-to-date on all aspects related to standardisation:​​​​

  • Discover the new regulatory developments;
  • Take part in developing standards;
  • Learn how to integrate standardisation in your R&D&i project;
DomainStandardStatusURL
Calidad del datoUNE 0079:2023PublishedSource
Gestión del datoUNE 0078:2023PublishedSource
Gobierno del datoUNE 0077:2023PublishedSource
Guía de evaluación de la Calidad de un Conjunto de DatosUNE 0081:2023PublishedSource
Guía de evaluación del Gobierno, Gestión y Gestión de la Calidad del DatoUNE 0080:2023PublishedSource
Medición del consumo energético, huella de carbono, consumo del agua y rendimiento de sistemas de Inteligencia ArtificialUNE 0086:2025PublishedSource

Additional translations in Spanish can be found here.

ISO/IEC Standards

DomainStandardStatusURL
AI Concepts and TerminologyISO/IEC 22989:2022 Information technology — Artificial intelligence — Artificial intelligence concepts and terminologyPublishedhttps://www.iso.org/standard/74296.html
AI ControllabitlityISO/IEC CD TS 8200 Information technology — Artificial intelligence — Controllability of automated artificial intelligence systemsPublishedhttps://www.iso.org/standard/83012.html
AI GovernanceISO/IEC 38507:2022 Information technology — Governance of IT — Governance implications of the use of artificial intelligence by organizationsPublishedhttps://www.iso.org/standard/56641.html
AI Management SystemISO/IEC DIS 42001 Information technology — Artificial intelligence — Management systemPublishedhttps://www.iso.org/standard/81230.html
AI Impact AssessmentISO/IEC 42005:2025 Information technology — Artificial intelligence (AI) — AI system impact assessmentPublishedhttps://www.iso.org/standard/42005
AI PerformanceISO/IEC TS 4213:2022 Information technology — Artificial intelligence — Assessment of machine learning classification performancePublishedhttps://www.iso.org/standard/79799.html
AI PrivacyISO/IEC AWI 27091 Cybersecurity and Privacy — Artificial Intelligence — Privacy protectionUnder Developmenthttps://www.iso.org/standard/56582.html
AI QualityISO/IEC AWI TR 42106 Information technology — Artificial intelligence — Overview of differentiated benchmarking of AI system quality characteristicsUnder Developmenthttps://www.iso.org/standard/86903.html
AI Risk ManagementISO/IEC 23894:2023 Information technology - Artificial intelligence - Guidance on risk managementPublishedhttps://www.iso.org/standard/77304.html
AI SecurityISO/IEC DIS 27090 Cybersecurity — Artificial Intelligence — Guidance for addressing security threats and failures in artificial intelligence systemsUnder Developmenthttps://www.iso.org/standard/56581.html
AI SustainabilityISO/IEC AWI TR 20226 Information technology — Artificial intelligence — Environmental sustainability aspects of AI systemsPublishedhttps://www.iso.org/standard/86177.html
AI Verification and ValidationISO/IEC AWI TS 17847 Information technology — Artificial intelligence — Verification and validation analysis of AI systemsUnder Developmenthttps://www.iso.org/standard/85072.html
AI Audit and CertificationISO/IEC 42006:2025 Information technology — Artificial intelligence — Requirements for bodies providing audit and certification of artificial intelligence management systemsPublishedhttps://www.iso.org/standard/42006
Biases in AIISO/IEC TR 24027:2021 Information technology — Artificial intelligence (AI) — Bias in AI systems and AI aided decision makingPublishedhttps://www.iso.org/standard/77607.html
Ethical and societal concernsISO/IEC TR 24368:2022 Information technology — Artificial intelligence — Overview of ethical and societal concernsPublishedhttps://www.iso.org/standard/78507.html
ExplainabilityISO/IEC AWI TS 6254 Information technology — Artificial intelligence — Objectives and approaches for explainability of ML models and AI systemsUnder Developmenthttps://www.iso.org/standard/82148.html
Biases in AIISO/IEC CD TS 12791 Information technology — Artificial intelligence — Treatment of unwanted bias in classification and regression machine learning tasksPublishedhttps://www.iso.org/standard/84110.html
Data Quality for AI/MLISO/IEC DIS 5259 Artificial intelligence — Data quality for analytics and machine learning (ML) (1 to 6)Publishedhttps://www.iso.org/standard/81088.html
Data LifecycleISO/IEC FDIS 8183 Information technology — Artificial intelligence — Data life cycle frameworkPublishedhttps://www.iso.org/standard/83002.html
TransparencyISO/IEC AWI 12792 Information technology — Artificial intelligence — Transparency taxonomy of AI systemsUnder Developmenthttps://www.iso.org/standard/84111.html
Trustworthy AIISO/IEC TR 24028:2020 Information technology — Artificial intelligence — Overview of trustworthiness in artificial intelligencePublishedhttps://www.iso.org/standard/77608.html
Synthetic DataISO/IEC AWI TR 42103 Information technology — Artificial intelligence — Overview of synthetic data in the context of AI systemsUnder Developmenthttps://www.iso.org/standard/86899.html
AI SafetyISO/IEC CD TR 5469 Artificial intelligence — Functional safety and AI systemsPublishedhttps://www.iso.org/standard/81283.html
Beneficial AI SystemsISO/IEC AWI TR 21221 Information technology – Artificial intelligence – Beneficial AI systemsUnder Developmenthttps://www.iso.org/standard/86690.html

Learning Resources for ISO/IEC Standards

  • ISO 42001 Visual Library — A visual learning library for ISO/IEC 42001:2023 covering all clauses, Annex A controls and the PDCA cycle through reference cards, memory cards and deep dives. Open source, CC BY 4.0.

NIST Publications

ResourceDescriptionSource
AI RMF (Risk Management Framework)The AI Risk Management Framework (AI RMF) is intended for voluntary use and to improve the ability to incorporate trustworthiness considerations into the design, development, use, and evaluation of AI products, services, and systems.Source
AI RMF PlaybookThe Playbook provides suggested actions for achieving the outcomes laid out in the AI Risk Management Framework (AI RMF) Core (Tables 1 – 4 in AI RMF 1.0). Suggestions are aligned to each sub-category within the four AI RMF functions (Govern, Map, Measure, Manage).Source
AI RMF GlossaryThis glossary seeks to promote a shared understanding and improve communication among individuals and organizations seeking to operationalize trustworthy and responsible AI through approaches such as the NIST AI Risk Management Framework (AI RMF).Source

Other resources

Additional standards can be found using the Standards Database and AIDG Hub, we recommend to review NIST Assessing Risks and Impacts of AI (ARIA), and EU AI Act Harmonised Standards Mapping as well. Another interesting repository for AI Governance is the AI Governance Library.

Tools

This section features tools and libraries that help to design, implement and manage AI in a responsible way.

ToolLanguageDescriptionCreatorStatus
balancePythonA package for balancing biased data samplesMETAActive
clavRThis package provides utilities for conducting cluster (profile) analysis with an emphasis on the validating the stability of the profiles both within a given data set as well as across data sets.Jason BryerActive
smclafifyPythonBias detection and mitigation for datasets and modelsAmazonInactive
SolasAIPythonA Library of Curated Disparity Testing Metrics for Use in Real-World SettingsSolasAIActive
TRAK (Attributing Model Behaviour at Scale)PythonA data attribution method called TRAK (Tracing with the Randomly-Projected After Kernel) to make accurate counterfactual predictions. See: ArticleMITInactive

This section is under review and the rest of entries will be added to the table with extended information.

AI Alignment

AI Governance

AI Licensing

Licensing AI models adds new layers of complexity beyond what traditional software licenses manage. Models may have separate licenses for: (1) The code used to train the model, (2) The model weights after training, (3) The datasets used during training, and (4) The outputs generated when users interact with the model.

Audit

  • AIR Blackbox Python - Open-source EU AI Act compliance scanner and runtime trust layer for Python AI agents. HMAC-SHA256 tamper-evident audit chains, PII detection, and prompt injection blocking. Trust layers for LangChain, CrewAI, AutoGen, OpenAI, Google ADK, and Claude Agent SDK. (Website | PyPI)
  • PRML / falsify Python, JS, Go, Rust Studio 11 - Pre-Registered ML Manifest specification (CC BY 4.0). Commits an evaluation claim (metric, comparator, threshold, dataset hash, seed, producer identity) to a SHA-256 hash before the experiment runs. Tamper-evident audit trail; subcategory crosswalks for EU AI Act Article 12, NIST AI RMF, ISO/IEC 42001. Four byte-equivalent reference implementations across 20 conformance vectors. Zenodo DOI 10.5281/zenodo.20177839, in SchemaStore catalog.
  • glassalpha Python
  • Systima Comply TypeScript Systima

Causal Inference

Data Management

Data Quality

Data Version Control

Drift

EU AI Act Compliance

Fairness

Feature Stores

Interpretability/Explicability

Interpretable Models

Model Verification

LLM Evaluations and Benchmarks

Measuring progress is fundamental to the advancement of any scientific field. As benchmarks play an increasingly central role, they also grow more susceptible to distortion. Read more about it in Sing, S., et al. (2025) The Leaderboard Illusion. arXiv preprint arXiv:2504.2087. In addition to the problem of distursion, we must remember that this is nascent discipline as stated in Weidinger, L., et al. (2025). Toward an evaluation science for generative AI systems. arXiv preprint arXiv:2503.05336. Benchmarks may appear as neutral scoreboards; however, they embody more than that. Each one signifies a particular philosophy: the types of work valued, the definition of success, and what can be safely disregarded. The development of a truly effective benchmark is equally challenging and indispensable as the development of the model itself.

New approaches are emerging to established protocol or methodology for conducting AI evaluations such as PREP-Eval v1.0 Pre-registration and REporting Protocol for AI Evaluations or Evals-Consensus.

Additional benchmarks can be found here, the AI Benchmarking Hub (from Epoch) compares the latest frontier AI models against each other, the CAIS AI Dashboard (from the Center for AI Safety) provides their latest benchmarks (text, vision, safety and automation), and you can learn about prompt evaluations here (by Anthropic).

LLM Regulation Compliance

  • COMPL-AI Python ETH Zurich Insait LaticeFlow AI
  • Tunix Python Google

Performance (& Automated ML)

(AI/Data) Poisoning

Privacy

Red Teaming

Reliability Evaluation (of post hoc explanation methods and LLMs evaluations)

Robustness

Safety

Security

For consumers:

Synthetic Data

Sustainability

(RAI) Toolkit

(AI) Watermarking

Citing this repository

Contributors with over 50 edits can be named coauthors in the citation of visible names. Otherwise, all contributors with fewer than 50 edits are included under "et al."

Bibtex

@misc{arai_repo,
  author={Josep Curto et al.},
  title={Awesome Responsible Artificial Intelligence},
  year={2026},
  note={\url{https://github.com/AthenaCore/AwesomeResponsibleAI}}
}

ACM, APA, Chicago, and MLA

ACM (Association for Computing Machinery)

Curto, J., et al. 2026. Awesome Responsible Artificial Intelligence. GitHub. https://github.com/AthenaCore/AwesomeResponsibleAI.

APA (American Psychological Association) 7th Edition

Curto, J., et al. (2026). Awesome Responsible Artificial Intelligence. GitHub. https://github.com/AthenaCore/AwesomeResponsibleAI.

Chicago Manual of Style 17th Edition

Curto, J., et al. "Awesome Responsible Artificial Intelligence." GitHub. Last modified 2026. https://github.com/AthenaCore/AwesomeResponsibleAI.

MLA (Modern Language Association) 9th Edition

Curto, J., et al. "Awesome Responsible Artificial Intelligence". GitHub, 2026, https://github.com/AthenaCore/AwesomeResponsibleAI. Accessed 28 May 2026.