Society & AI
Research & Commentary Advisors About

Research Ethics Statement

This Research Ethics Statement was last updated on October 5, 2025. It is informed by principles from the Belmont Report, the American Educational Research Association Code of Ethics, and participatory research traditions worldwide.

Effective Date: October 5, 2025
Last Updated: October 5, 2025

Our Commitment

The Society & AI Independent Research Group (Society & AI IRG) is committed to conducting research that adheres to the highest ethical standards. This Research Ethics Statement articulates the principles that guide our scholarly work, community engagement, and public communication.

Core Ethical Principles

1. Independence and Integrity

Research Autonomy: Our research agenda is determined by scholarly merit, educational need, and social value, and never by commercial interests, political pressures, or funding contingencies.

Intellectual Honesty: We report findings accurately, acknowledge limitations transparently, and correct errors promptly. We do not suppress inconvenient results or overstate conclusions.

Conflict of Interest: We disclose any financial, professional, or personal relationships that could reasonably be perceived as influencing our research. Where conflicts exist, we recuse ourselves from relevant decisions.

2. Respect for Persons and Communities

Informed Participation: When research involves human subjects or community data, we obtain meaningful informed consent. Participants understand what is being studied, potential risks, and their right to withdraw.

Privacy and Confidentiality: We protect the privacy of research participants and communities. Personal data is anonymized, stored securely, and used only for stated purposes.

Cultural Respect: We honor diverse epistemologies, cultural protocols, and community knowledge systems. We do not extract knowledge without reciprocal benefit or impose Western frameworks as universal standards.

Voice and Agency: Communities and educators are not merely research subjects but partners with genuine decision-making authority over how research is conducted and disseminated.

3. Beneficence and Non-Maleficence

Public Benefit: Our research aims to advance educational equity, social justice, and human flourishing. We prioritize questions that serve communities facing systemic marginalization.

Do No Harm: We assess potential harms—to individuals, communities, and the knowledge commons—before beginning research. When harms are possible, we implement safeguards or reconsider the study.

Vulnerable Populations: We exercise heightened care when research involves children, marginalized communities, or individuals with limited institutional power.

4. Justice and Equity

Inclusive Participation: We actively recruit diverse perspectives in our research processes, advisory roles, and authorship. We do not center elite institutions or Global North contexts as default.

Equitable Benefit Sharing: Communities whose knowledge, data, or participation contribute to research receive tangible benefits—whether through capacity building, shared intellectual property, policy advocacy, or resource allocation.

Accessible Scholarship: We publish under open-access licenses whenever possible. Knowledge generated through public-interest research should return to the public, not be locked behind paywalls.

Specific Ethical Commitments

Research Design

  • Participatory Methods: We employ co-design, community-based participatory research, and deliberative methods that share power between researchers and communities.
  • Methodological Pluralism: We use multiple methods—quantitative, qualitative, design-based—to capture complexity and avoid reductionism.
  • Transparency: We publish preregistrations, protocols, and data (where ethically permissible) to enable scrutiny and replication.

Data Governance

  • Minimal Collection: We collect only data necessary for stated research purposes.
  • Secure Storage: Data is encrypted, access-controlled, and stored in compliance with applicable regulations.
  • Data Sovereignty: For research involving Indigenous knowledge or community data, governance follows community protocols and respects data sovereignty principles.
  • Retention and Deletion: We retain data only as long as necessary and delete it when no longer needed or upon participant request.

AI and Technology Ethics

Given our focus on AI in education, we hold ourselves to additional standards:

  • Algorithmic Accountability: When we develop or evaluate AI systems, we document training data, model limitations, and known biases.
  • Fairness Testing: We test systems across demographic groups to identify disparate impacts.
  • Human Oversight: We oppose automation of high-stakes educational decisions without meaningful human review.
  • Avoiding Harm: We refuse to build or promote AI systems that surveil students, manipulate behavior, or undermine teacher autonomy.

Authorship and Credit

  • Fair Attribution: Authorship reflects substantial intellectual contribution. We acknowledge all contributors appropriately.
  • No Ghost Authorship: Listed authors have participated meaningfully; those who have not are not included.
  • Student and Junior Scholars: We mentor early-career researchers fairly and credit their contributions appropriately.
  • Community Knowledge: When research builds on community expertise or Indigenous knowledge, we acknowledge this explicitly and obtain permission for use.

Peer Review and Publication

  • Constructive Review: When reviewing others’ work, we provide fair, substantive feedback focused on improving scholarship.
  • Confidentiality: We do not misuse privileged information from peer review.
  • No Self-Plagiarism: We do not republish our own work without disclosure.
  • Retraction: If we discover significant errors, we issue corrections or retractions as warranted.

Public Communication

  • Accuracy: We communicate research findings accurately to public audiences, avoiding overstatement or sensationalism.
  • Accessible Language: We translate technical findings into language accessible to educators, policymakers, and communities.
  • Context: We provide appropriate context and caveats so findings are not misinterpreted or misused.

Accountability Mechanisms

Internal Review

Research protocols undergo internal ethical review before initiation. High-risk studies are reviewed by external ethics boards or institutional review boards (IRBs).

Advisory Oversight

Our Community Advisory Board includes volunteer educators, researchers, and community representatives who review research priorities and ethical practices from time to time.

Grievance Process

Individuals or communities who believe research has caused harm or violated ethical principles may file grievances to the Society & AI Principal Researcher, Sai Gattupalli, Ph.D., at sai@societyandai.org. Complaints are reviewed promptly, and we commit to transparent resolution.

Continuous Improvement

We review and update this Ethics Statement annually, incorporating feedback from collaborators, participants, and the scholarly community.

Limitations and Humility

We acknowledge that:

  • Ethical practice is evolving. Standards that seem adequate today may prove insufficient tomorrow.
  • Power imbalances persist. Despite best efforts, academic research often reproduces inequities.
  • We will make mistakes. When we do, we commit to transparency, accountability, and learning.

Contact

For questions about research ethics or to report concerns:

Society & AI Independent Research Group (SAI IRG)
Principal Researcher Contact: sai@societyandai.org
Website: https://societyandai.org