AI summary by TwinMind
If AI can do your classwork, why go to college? | The Gray Area
Vox
49m 34s
Introduction and Background
- Sean (host of The Gray Area) interviews James Walsh, features writer at New York Magazine
- Discussion of Walsh's piece about AI cheating in college and threats to academic integrity
- Walsh initially started investigating general cheating but realized "no way to write a cheating story without it being about AI"
- Found "everybody I'm talking to is in some way cheating, even if they don't think it's cheating"
Traditional vs. AI-Enabled Cheating
- Traditional cheating methods still exist (Apple Watches for PDFs, scribbling on palms)
- Student quote: "the floor of cheating is still there. It's just that like the ceiling's been blown off"
- AI has fundamentally changed scale and accessibility of academic dishonesty
Student Interview Process and Candor
- Found students through student newspapers, op-eds they had written
- Notable example: freshman wrote optimistic AI op-ed, then as sophomore said "I have done a 180. We need to get AI out of the classroom"
- Reddit/Discord outreach - Canadian student posted about "chat GPT addiction" seeking help
- Students surprisingly open about their AI usage
AI Platforms and Usage Patterns
- ChatGPT most commonly used (often used as shorthand for all AI chatbots)
- Also using Claude (Anthropic), Gemini (Google), Copilot
- Usage across all academic facets:
- Essay writing, outlining, idea generation
- Note-taking during class
- Textbook summarization
- Study guide creation
- Computer science students using for coding
- Science students for data analysis
Computer Science Specific Issues
- Berkeley CS professor reports "serious problem" with students using AI on take-home assignments
- Students failing in-class tests on same material they used AI for at home
- Professor warns students: "if you are depending on AI now you are just training yourself to be an assistant to an AI platform"
- Tech executives asking why they would hire coders now
- Microsoft recent layoffs of coders mentioned
Student Perspectives and Motivations
- Whiplash effect: uncertainty about what skills will be valuable in 4 years
- "I might as well offload this learning...because who knows if four years from now they're going to help me"
- Three main student categories:
- Those recognizing AI as problematic (Wisconsin student noticing library full of ChatGPT users)
- Those feeling compelled to use AI to stay competitive
- Those viewing it as efficiency tool for "hackable assignments"
Simple Cheating Process
- Basic method: copy professor's prompt, paste into ChatGPT, request 4-5 page essay
- Many students copying/pasting without reading output
- "Trojan horse" method by professors: embedding nonsensical instructions (mention broccoli, Dua Lipa, Finland)
- Students blindly submitting essays containing these markers
Detection Challenges
- AI detectors assign probability percentages (e.g., "70% chance this is AI generated")
- Only analyze language patterns, not ideas or critical thinking
- Don't catch students using AI for ideation then writing themselves
- Professors know many more students using AI than they can prove
Roy Lee Case Study
- Community college to Columbia transfer, startup founder aspirant
- Used AI to "cruise through every assignment" at Columbia
- Attended elite school "to find a co-founder and a wife"
- Represents transactional view of college education
- Example of college as networking rather than learning
Wendy's Perspective - Redefining Cheating
- Distinguished between copying/pasting (which she considered cheating) vs. using AI for ideation
- Used ChatGPT for topic sentences, formulating ideas - "paint by numbers essay"
- Could complete in 2 hours what normally took 6-8 hours
- Nostalgic for actual writing but didn't consider her AI usage as cheating
- Wrote about critical pedagogy using AI assistance - didn't initially see irony
Professor Responses and Despair
- Primarily writing-heavy and computer science professors in despair
- Some AI-optimistic professors exist (UCLA comparative literature professor created entire textbook with AI)
- Most professors don't know how to police AI usage
- Difficulty proving AI usage even when suspected
- Volume of AI-generated work overwhelming
- Many professors considering retirement rather than adapting
Detection and Enforcement Issues
- "CSI sort of thing" for every AI essay unsustainable for professors
- Focus shifting to convincing students why they shouldn't use AI
- Practical adaptations: return to blue books, oral exams
- One professor switched entirely to oral exams, enjoys one-on-one time
- Acknowledges something lost in transition from written to oral assessment
Faculty AI Usage
- Professors increasingly using AI for lectures, textbooks, lesson planning
- Platforms advertising to professors for grading papers
- "Very real possibility that right now an AI is grading itself" - AI grading AI-written essays
- Microsoft providing AI platforms to entire school networks (São Paulo example)
- Students catching professors using AI
Administrative Response
- Many administrators showing AI optimism, wanting to be part of "new wave"
- Professors feeling "very much on their own" with AI challenges
- Comparison to COVID response: quick institutional changes for financial threat vs. slower AI response
- AI doesn't threaten bottom line immediately like COVID did
- Universities comfortable with "degraded education as long as tuition checks are still cashing"
Systemic Educational Issues
- AI exposing existing "rot underneath education"
- Higher education as transactional experience for decades
- "Pay X amount of dollars, tens of thousands of dollars, and you get your degree. What happens in between is not as important"
- University-AI company partnerships for customer capture
- OpenAI reportedly sitting on effective AI detector but not releasing it
Personal Impact on Writing and Thinking
- Walsh admits he would have used AI as 18-year-old despite hating writing
- Recognition that writing is thinking - "I don't even know what I think until I write"
- Concern about students missing struggle from "blank blinking cursor to one page"
- Fear of losing "sense of self" and "ability to think in complex critical ways"
Technology Comparison Debate
- AI optimists compare to calculator, spellcheck evolution
- Walsh emphasizes fundamental difference between spellcheck corrections and AI generating ideas/summarizing books
- Not just "old people yelling at clouds" - tool available to all generations
- Sam Altman's "calculator for words" statement contrasted with warnings about AI being "lights out for humanity"
Post-Literate Society Concerns
- Fear of "post-literate world" where fewer people read and think
- Writer's concern that "most important reader now is a robot"
- Technology's demonstrated impact on teen brains over past decade
- AI orders of magnitude more powerful than previous algorithms
- No model of liberal democracy that works in post-literate society
Educational Inequality Implications
- AI potential benefits: helpful for second/third language writers, customized study guides
- Risk of writing becoming "anachronistic elective like basket weaving"
- Only privileged students at certain schools would learn to write and think
- Question of who will "afford to have those experiences"
