Faculty: Dorris Scott, School of Continuing and Professional Studies

Course: Data Fundamentals I (DATA 200)

TL;DR

Students test a project dataset against data ethics and privacy standards, then use a university-approved AI tool to check it against FAIR principles (Findable, Accessible, Interoperable, and Reusable), feeding the AI both their data and key institutional guidelines. A manual human audit closes the loop, catching any AI errors or privacy leaks before the data is shared.

What is being done

In Data Fundamentals I, students learn core data concepts such as data literacy, data management, data analysis and data ethics to prepare raw information for real-world decision making. 

Before AI enters the picture, students first evaluate their dataset on their own, checking for human subjects risk, unclear ownership, and sharing restrictions tied to the project they’re building toward. Only after completing this manual ethics review do they bring in generative AI, working through a three-stage workflow using a privacy-protected, university-approved tool selected from WashU’s AI Tool Comparison Chart:

  1. Baseline Evaluation: Students upload their raw dataset and prompt the AI to flag ethics concerns and suggest ways to align the dataset with FAIR principles.
  2. Multi-Document Evaluation: Students deepen the AI’s analysis by feeding it content from the course and  institutional data management materials and guidance from the WashU Libraries Data Sharing Research Guide alongside their dataset. Prompted to cross-reference all of these resources, the AI scans for hidden personally identifiable information (PII) and generates a cleaned, FAIR-aligned sample of the dataset. 
  3. Human-in-the-Loop Override: Students independently audit the AI’s output, documenting anywhere it misread a guideline, invented a data point, or missed sensitive information, then manually correct the dataset to produce a clean CSV file and an updated README file.

This pipeline adds value by turning abstract data ethics principles into a hands-on audit. Rather than asking AI to produce finished work, the assignment teaches students to use AI to handle tedious formatting and initial compliance checks, while training themselves to catch algorithmic errors, a vital skill for future data practitioners.

Adapting Across Disciplines

This structure, an AI tool drafting against institutional guidelines followed by a mandatory human audit, transfers easily to other field:

  • Public Health & Clinical Research: Students could process de-identified patient records alongside HIPAA guidelines, prompting AI to catch remaining demographic identifiers before building a final data dictionary.
  • Digital Humanities: Students could upload historic census logs or archival sheets alongside metadata standards, letting AI standardize location names and spellings, then writing a log documenting anywhere the AI flattened historical or regional nuance.
  • Environmental & Climate Sciences: Students could feed raw weather or sensor logs into an AI tool alongside formatting rules to flag outliers, then manually confirm the results meet scientific archiving standards.

Generative AI Teaching Repository

The assignment was developed through the AI Curriculum Corps in partnership with the Center for Teaching and Learning. The CTL Generative AI Teaching Repository serves as a centralized hub for WashU faculty to share, discover, and adapt innovative AI-integrated assignments. To explore peer-reviewed class activities or contribute your own framework, visit the CTL Generative AI Teaching Repository.

Join the AI Curriculum Corps

Looking to responsibly integrate AI into your curriculum? The AICC provides WashU faculty with technical support, peer feedback, and guidance on navigating secure university-supported AI tools. To apply for the next cohort of the AICC, visit ai.washu.edu/curriculum