AI Ethics: Privacy, Security, and Bias
[Unedited image created by AI. Errors maintained for demonstration purposes.]
Overview
GenAI faces significant ethical challenges, including biases in training data that can lead to discriminatory outputs and issues with content moderation. AI is increasingly being used in critical areas like healthcare and decision-making, which can result in biased or harmful outcomes. Additionally, AI’s environmental impact is substantial, contributing to high levels of carbon emissions and water consumption. Tools covered include Zotero, Consensus, and Elicit.
Videos
AI Ethics
Zotero
Consensus and Elicit
Suggested Readings
- Covert Racism in LLMs (Overview, full paper linked)
- John Oliver on AI
- $2 Workers that Made AI Safe
Suggested Assignment
AI Ethics Literature Review on AI in Academic Writing
Overview:
Objectives:
- Critically examine and synthesize existing research on the use of AI in academic writing.
- Evaluate the ethical implications of using AI tools for academic purposes.
- Understand and apply Fitchburg State University’s academic integrity policy to your analysis.
- Develop a well-supported argument based on your review of the literature.
Instructions:
-
Literature Search and Analysis:
- Use Elicit to gather scholarly articles and papers on AI ethics, with a focus on its use in academic writing.
- Apply Consensus to synthesize findings and discern major themes and ethical considerations.
-
Writing the Literature Review:
- Your review should specifically address the ethical implications of using AI in academic settings.
- Incorporate a discussion on Fitchburg State University’s academic integrity policy, analyzing how it relates to the use of AI tools.
- Organize the literature into themes or categories that support your analysis and conclusion regarding the ethics of AI in academic writing.
-
Citation and Referencing:
- Cite all sources accurately using a citation style approved by your department.
- Questions to Consider (from AI and Writing):
- Some claim that every GenAI output is inherently plagiarism, on grounds that everything the GenAI creates was originally someone else’s words or thoughts. Yet, similarly, most of what we say, write, and think could be aruged to be plagiarism, because we so often cull it, reorder it, and re-present it from other peoples’ ideas and expressions. First, consider the ramifications of this on how we think about things like individuality, sense of self, expression, creativity, and even intelligence. Second, consider how such an acknowledgment of the similarities between GenAI and human thinking might affect what we understand about plagiarism and academic integrity.
- When you write, it’s likely that you already use some AI tools such as spell checkers and grammar checkers. These tools have altered our learning in many ways. You no longer need to know how to spell every word correctly. You no longer need to know all of the rules of grammar. Is using these kinds of tools a violation of academic integrity? Write a short essay comparing the accepted use of AI tools such as spelling and grammar checkers with the often-prohibited use of GenAI.
- Is monitoring for plagiarism an act of policing or an act of education?
Submission Requirements:
- A literature review document of 1500-2000 words addressing the specified topic.
- A bibliography listing all referenced sources.
Rubric for AI Ethics Literature Review on AI in Academic Writing
Criteria | Excellent (90-100%) | Good (80-89%) | Satisfactory (70-79%) | Needs Improvement (<70%) |
---|---|---|---|---|
Relevance and Depth of Research | Exceptionally relevant sources are used; provides deep insight into ethical considerations of AI in academic writing. | Sources are relevant and provide good coverage of the topic with substantial insight. | Sources cover the topic adequately but lack depth or broader insight. | Sources are insufficiently relevant or fail to adequately cover the topic. |
Integration of FSU Academic Integrity Policy | Excellently integrates FSU’s academic integrity policy, providing a strong ethical framework for the analysis. | Adequately integrates the academic integrity policy with good relevance to AI ethics. | Mentions FSU’s policy but with limited integration or relevance to AI ethics. | Fails to effectively incorporate or relate FSU’s academic integrity policy. |
Use of AI Tools | Effectively uses AI tools to enhance the breadth and depth of literature analysis. | Uses AI tools well, contributing positively to the analysis. | Uses AI tools adequately, but integration into research is basic. | Uses AI tools minimally or ineffectively; does not enhance research. |
Organization and Clarity | Review is exceptionally well-organized and clearly presents a coherent narrative regarding the ethics of AI in academic writing. | Well-organized and presents clear findings with minor issues. | Somewhat organized but lacks clarity in presenting findings. | Poorly organized and difficult to follow. |
Critical Analysis and Argumentation | Provides a deep, insightful analysis with a strong, well-supported argument regarding the ethical use of AI. | Solid analysis and argumentation with some insightful observations. | Basic analysis with an adequate argument but lacks depth. | Lacks critical analysis or fails to form a coherent argument. |
Citation and Referencing | All sources are cited flawlessly according to the recommended citation style. | Minor errors in citation style but generally well done. | Citation style is inconsistent or has several errors. | Many citation errors or incorrect use of citation style. |