Overview
This training workshop equips researchers with the essential skills required to apply ethical practices in AI to their respective fields of study and integrate these methodologies into their research. The overarching objective is to foster a research community of researchers that have a foundational knowledge of the principles of ethical AI through interactive, case-based learning and peer discussion.
Topics
- Foundations of Research Ethics
- Privacy and Surveillance
- Algorithmic Bias
- AI Safety & Security
- Transparency, Explainability, and Accountability
- AI Impact on Society
- AI Policy and Regulations
Additional Information
By the end of the workshop, participants will be able to:
- Identify and analyze ethical impact of their own research contexts
- Apply responsible data practices to research design and implementation
- Integrate ethical considerations into grant proposals and research protocols
- Evaluate AI systems as informed citizens and consumers
- Engage productively with diverse stakeholders about AI ethics
This course is offered free.
U-M Researchers (faculty, staff, postdocs, and graduate students). Priority registration offered to postdoctoral fellows and graduate students.
This is designed for researchers in a wide range of research domains, including biological and biomedical sciences, engineering, environmental and earth sciences, physical sciences, and social sciences.
Suggest reviewing Responsible Generative AI Specialization on Coursera (4 asynchronous Modules – 8 hours)
In-person, all-day small group session. Lunch is provided.
Jan 11-12, 2027: Two-day workshop followed by 3 hands-on labs
Participants who complete the workshop and provide feedback through a survey will receive a training certificate of completion and digital training badge.
Course Instructor

Merve Hickok
Responsible Data and AI Advisor
Lab Instructors

Ali Bolcakan
Postdoctoral Affiliates Fellow, 2025 Cohort

Nathan Fox
Schmidt AI in Science, 2022 Cohort, Alum

Xiaosu (Frank) Hu
Data Scientist
Workshop Schedule
Day 1 – January 11
9:00-10:30 AM:
Foundations of Research Ethics
- Historical context: Tuskegee, Henrietta Lacks, algorithmic harm cases
- Reproducibility in AI research
Exercise: Discuss how participants currently examine ethical dimensions in their own research projects, and what is their understanding of ethics
10:45 AM-12:15 PM
Privacy and Surveillance
- Privacy and surveillance risks
- Power of data – unintended consequences of data
- De-identification, re-identification, aggregation risks
- Limitations of consent
Exercise: Discussion of experiences with AI as consumers
12:15-1:00 PM
Lunch
1:00-2:15 PM
Algorithmic Bias
- Types and Sources of bias (historical, representation, measurement, evaluation)
- Real-life incidents
Exercise: Identify potential bias sources in research
2:30-3:30 PM:
AI Safety & Security
- Dimensions of AI safety (technical robustness, social/ethical, legal)
- Hallucination
- Security (cybersecurity, data protection)
- Reputational harms (liability, defamation)
- Child safety
- Public safety
Exercise: Discuss impact of hallucination in academic research process
Day 2 – January 12
Exercise: Create documentation for a research dataset or model
10:45 AM – 12:15 PM
Module 6: AI Impact on Society
- Deepfake deception (fraud, manipulation, disinformation)
- Disinformation & Manipulation
- Consumer protection
- Human right implications
- Digital divide and access inequities
- Social and cultural impact (erosion of trust, copyright, stereotyping, homogenization)
Exercise: Stakeholder mapping for participant research projects
12:15 PM – 1:00 PM
Lunch
1:00 PM – 3:00 PM
Module 7: AI Policy and Regulations:
- Rights regarding personal data and AI decisions; existing legal and regulatory protections
- Global AI regulation landscape (EU, US, California)
3:00 PM – 3:45 PM
Lab Session Introduction and Sign-up:
Hands-On Labs (1 hour, led by MIDAS staff scientists – in the Sandbox each week after the workshop)
- Lab 1: Bias Detection and Mitigation
- Lab 2: LLM Leaderboards
- Lab 3: Explainability
TBD
Project: Integrate learning into participants’ actual research contexts
- Session 1: Interdisciplinary peer discussion about current/hypothetical research scenario
- Session 2 – N: Assessment of current/hypothetical research scenario, with stakeholder analysis, ethical risk assessment, mitigation strategies, documentation approach
Questions?
Reach out to [email protected]