On this page, you will read about vulnerabilities in assignment requirements (in relation to GenAI) that can often be fixed. You will also find information on solutions and detailed examples. There are explanations of four specific vulnerabilities, each of which is detailed on a separate page.

By clicking on the links you can view the other pages or go back to the page on written assignment without supervision.

The assignment does not require course-specific knowledge
The assignment does not require higher-level skills
The assignment does not require students to engage with discipline-specific literature
The assignment does not require visualizations

ChatGPT is not very good at certain higher-level skills, such as contextual thinking, temporal and psychological reasoning, or interpretation. You can use this to your advantage when designing assessments. Include in your assignment instructions questions that ask students to show their understanding of the material in an authentic and meaningful way through (critical) reflection, analysis, interpretation, integration, evaluation, or synthesis.

How can you promote student reflection and creativity in assessment?

Have students make connections between several ideas or concepts;
Ask students to evaluate an argument or perspective;
Ask students to critically reflect on a specific academic paper;
Have students apply knowledge in a new situation or context, or on aspects of the course, as demonstrated in a particular class;

There are certain tasks that GenAI performs (very) well and others with which it struggles. As a rule of thumb, we can say that the simpler the task (e.g., summarize, describe, explain), the better the generated output. Instead, ask students to analyze, integrate, apply, and/or provide real-life examples.

Example: summary task

Assignment instructions

Write a short (250-word) summary of the role of Cambridge Analytica on the US election 2016.

Analysis

As we can see from the output, ChatGPT fares reasonably well on this assignment. This can be expected as the Cambridge Analytica scandal would be well represented in the training data for ChatGPT and the assignment “write a short summary” is relatively simple.

Student

Write a short (250-word) summary of the role of Cambridge Analytica on the US election 2016.

ChatGPT

Cambridge Analytica, a British political consulting firm, played a highly controversial role in the 2016 U.S. Presidential Election. The company utilized advanced data mining and analysis techniques to construct detailed psychological profiles of American voters. Cambridge Analytica leveraged its vast database – allegedly containing personal data of up to 87 million Facebook users obtained without their consent – to micro-target highly personalized political advertisements.

The firm was hired by Donald Trump’s election campaign, using its capabilities to influence voter behavior in favor of Trump. The sophistication of this data-driven campaign raised significant concerns about privacy, consent, and the potential for manipulation within the democratic process. Critics argued that the firm exploited the digital footprint of millions of unsuspecting Facebook users, capitalizing on their likes, shares, and other activities to sway the election.

Although it’s hard to quantify the direct impact Cambridge Analytica had on the election’s outcome, the revelations of its practices sparked a global conversation about data privacy and the role of social media in political campaigns. The scandal prompted significant backlash against Facebook and led to the eventual dissolution of Cambridge Analytica in 2018.

Example: writing an introduction

When we ask ChatGPT to write an entire introduction for a research paper on the same topic, it struggles to generate a satisfactory result.

Assignment instructions

Write an introduction for an academic paper investigating the role of microtargeting on social media in US elections.

Analysis

Even though the intro is well-written, it would not meet expectations for a research paper as it is too general, does not introduce a specific angle, and makes up an empirical analysis.

The main reason for the quality discrepancy in both tasks is that writing an introduction requires complex reasoning (Which concepts are important? How are they defined? What is a potentially interesting and under-researched RQ? Which analytical lens makes most sense to investigate the topic?).

Student

Write an introduction for an academic paper investigating the role of microtargeting on social media in US elections.

ChatGPT

Elections and their outcomes are fundamentally linked to the flow of information, and nowhere is that flow more pervasive or influential than on social media platforms. In recent years, the granularity of data available on individuals has led to the emergence of microtargeting as a primary strategy in political campaigns, particularly in the United States. Microtargeting leverages personal data to tailor political messages for small, specific groups or even individual voters. However, the implications of this tactic for democratic processes remain under-researched.

This paper investigates the role and impact of microtargeting on social media in U.S. elections. Through a multi-faceted lens, we explore the evolution of microtargeting, the data sources that facilitate it, and its application in the political arena. Drawing on data from several recent election cycles, we probe how these techniques have influenced voting behavior and the broader political discourse.

Furthermore, this research explores the ethical and democratic implications of this increasingly refined method of voter persuasion. Amid growing concern over data privacy, misinformation, and the potential manipulation of public sentiment, our study raises pertinent questions: To what extent does microtargeting empower or disenfranchise voters? Does it contribute to political polarization and echo chamber effects?

By charting this new frontier of digital politicking, this paper offers insights to academics, policymakers, election stakeholders, and citizens alike. In doing so, we aim to foster a broader understanding and discussion about the future of political engagement in the digital age. Our findings underscore the pressing need for regulatory scrutiny and potential reforms to ensure the fairness and integrity of future elections in the era of microtargeting.