2026-2027 Hands-On for AI Research Team
1 session on October 8, 2026
Description:
This session focuses on the early stages of working with data, when a dataset is large, complex, poorly documented, or unfamiliar. Participants will use AI to interpret data dictionaries, surface undocumented assumptions, and assess whether a dataset can answer their research question. They will also learn how to determine which AI tools they are permitted to use with a given dataset, since research data frequently carries terms of use that restrict AI tools. Participant exercises use a public-domain dataset and U-M's own AI services.
- ICPSR policy on the use of large language models: https://www.icpsr.umich.edu/sites/icpsr/about/policies/large-language-models-and-ai
- ICPSR redistribution policy: https://www.icpsr.umich.edu/sites/icpsr/about/policies/redistribution
- Using AI with MIDUS, NSHAP, Add Health and other NACDA-hosted data: https://www.icpsr.umich.edu/sites/icpsr/news/using-ai-with-midus-nshap-add-health-and-other-nacda-hosted-data
- U-M guidance on AI and U-M data: https://safecomputing.umich.edu/protect-the-u/safely-use-sensitive-data/AI-and-UM-Data
- U-M generative AI services: https://genai.umich.edu/
- ACS PUMS documentation: https://www.census.gov/programs-surveys/acs/microdata/documentation.html
Possible hands-on activities:
- Determine which tier of AI tool you may use with a dataset you are currently working with, and where to check
- Translate a cryptic codebook into plain language, with the model flagging what it cannot determine instead of guessing
- Extract documentation into a structured format that records, field by field, what the model could not determine from the documentation alone
- Generate a fitness-for-purpose checklist against your own research question, and draft specific questions to send a data provider
Before the session:
1. All attendees please bring a laptop.
2. Sign in at genai.umich.edu to make sure your U-M GPT access is working. We'll use it for all hands-on exercises.
3. Come with a research question you're working on, and the name of a dataset you're using or considering for it. If the dataset has a data use agreement or terms of use page, have the link handy. Please don't plan to upload your own data during the session. We'll use a public dataset for the exercises, and part of the session covers how to check which AI tools you're permitted to use with your own data.
This session focuses on the early stages of working with data, when a dataset is large, complex, poorly documented, or unfamiliar. Participants will use AI to interpret data dictionaries, surface undocumented assumptions, and assess whether a dataset can answer their research question. They will also learn how to determine which AI tools they are permitted to use with a given dataset, since research data frequently carries terms of use that restrict AI tools. Participant exercises use a public-domain dataset and U-M's own AI services.
- ICPSR policy on the use of large language models: https://www.icpsr.umich.edu/sites/icpsr/about/policies/large-language-models-and-ai
- ICPSR redistribution policy: https://www.icpsr.umich.edu/sites/icpsr/about/policies/redistribution
- Using AI with MIDUS, NSHAP, Add Health and other NACDA-hosted data: https://www.icpsr.umich.edu/sites/icpsr/news/using-ai-with-midus-nshap-add-health-and-other-nacda-hosted-data
- U-M guidance on AI and U-M data: https://safecomputing.umich.edu/protect-the-u/safely-use-sensitive-data/AI-and-UM-Data
- U-M generative AI services: https://genai.umich.edu/
- ACS PUMS documentation: https://www.census.gov/programs-surveys/acs/microdata/documentation.html
Possible hands-on activities:
- Determine which tier of AI tool you may use with a dataset you are currently working with, and where to check
- Translate a cryptic codebook into plain language, with the model flagging what it cannot determine instead of guessing
- Extract documentation into a structured format that records, field by field, what the model could not determine from the documentation alone
- Generate a fitness-for-purpose checklist against your own research question, and draft specific questions to send a data provider
Before the session:
1. All attendees please bring a laptop.
2. Sign in at genai.umich.edu to make sure your U-M GPT access is working. We'll use it for all hands-on exercises.
3. Come with a research question you're working on, and the name of a dataset you're using or considering for it. If the dataset has a data use agreement or terms of use page, have the link handy. Please don't plan to upload your own data during the session. We'll use a public dataset for the exercises, and part of the session covers how to check which AI tools you're permitted to use with your own data.
1 session on November 9, 2026
Focus: Using AI to support data preparation and quality improvement.
Description:
This session would examine how AI can help researchers think through data cleaning tasks before formal analysis begins. The emphasis would be on using AI as a support tool to identify inconsistencies, plan, clean workflows, document decisions, and generate reusable code templates, while keeping human review central.
Possible hands-on activities:
Identify common data quality issues in a sample datasetDraft a cleaning plan for missing values, duplicates, or formatting inconsistenciesGenerate starter code for basic cleaning steps in Python or RCreate a data cleaning log or documentation templateCompare raw and cleaned data examples to assess impactReminder: All attendees, please bring a laptop.
Description:
This session would examine how AI can help researchers think through data cleaning tasks before formal analysis begins. The emphasis would be on using AI as a support tool to identify inconsistencies, plan, clean workflows, document decisions, and generate reusable code templates, while keeping human review central.
Possible hands-on activities:
Identify common data quality issues in a sample datasetDraft a cleaning plan for missing values, duplicates, or formatting inconsistenciesGenerate starter code for basic cleaning steps in Python or RCreate a data cleaning log or documentation templateCompare raw and cleaned data examples to assess impactReminder: All attendees, please bring a laptop.
1 session on February 4, 2027
Description:
Explore how generative AI can support a rigorous, advanced approach to pre-registration. This session will demonstrate how researchers can use AI to develop analysis code and create simulated datasets before collecting real-world data—helping them clarify assumptions, test analytical workflows, identify potential problems, and strengthen research transparency and reproducibility.
Explore how generative AI can support a rigorous, advanced approach to pre-registration. This session will demonstrate how researchers can use AI to develop analysis code and create simulated datasets before collecting real-world data—helping them clarify assumptions, test analytical workflows, identify potential problems, and strengthen research transparency and reproducibility.
1 session on March 24, 2027
Focus: Using AI to critically assess outputs, check reasoning, and strengthen trust in findings.
Description:
This session would focus on using AI as a tool for review rather than solely for generation. Participants would practice checking results for plausibility, identifying possible errors or overclaims, and building habits for validation and reproducibility. This is especially important for avoiding misplaced confidence in AI-assisted work.
Possible hands-on activities:
Ask AI to critique an interpretation of findingsCross-check whether a result matches the stated method or assumptionsGenerate a validation checklist for an analysis workflowIdentify possible sources of bias, confounding, or errorReview examples of flawed outputs and discuss how to catch them
Description:
This session would focus on using AI as a tool for review rather than solely for generation. Participants would practice checking results for plausibility, identifying possible errors or overclaims, and building habits for validation and reproducibility. This is especially important for avoiding misplaced confidence in AI-assisted work.
Possible hands-on activities:
Ask AI to critique an interpretation of findingsCross-check whether a result matches the stated method or assumptionsGenerate a validation checklist for an analysis workflowIdentify possible sources of bias, confounding, or errorReview examples of flawed outputs and discuss how to catch them
1 session on April 8, 2027
Focus: Using AI to communicate research clearly to different audiences.
Description:
This session would explore how AI can help researchers translate technical work into accessible and audience-appropriate communication. Participants would practice adapting the same core findings for different settings, such as collaborators, community partners, students, leadership, or public-facing summaries.
Possible hands-on activities:
Turn a technical summary into plain-language prose
Draft a short abstract, presentation blurb, or project update
Reframe the same findings for academic vs. non-academic audiences
Generate slide outlines or talking points from key results
Create a first draft of documentation, FAQs, or knowledge-transfer notes
Description:
This session would explore how AI can help researchers translate technical work into accessible and audience-appropriate communication. Participants would practice adapting the same core findings for different settings, such as collaborators, community partners, students, leadership, or public-facing summaries.
Possible hands-on activities:
Turn a technical summary into plain-language prose
Draft a short abstract, presentation blurb, or project update
Reframe the same findings for academic vs. non-academic audiences
Generate slide outlines or talking points from key results
Create a first draft of documentation, FAQs, or knowledge-transfer notes