Does sharing a headline or an article about how ChatGPT might make you dumb—without actually reading the article or investigating the research methodology—confirm we’re all pretty dumb even without ChatGPT?
Jenny and Greg Swan are back with a brand-new episode that dissects the MIT “Your Brain on ChatGPT” study that hijacked every timeline in the last week. Spoiler: the study’s sample size is smaller than a freshman dorm, hasn’t been peer-reviewed, and might be less about AI brain melt and more about über-clickable fear bait.
But don’t worry! Your brain cells (and ours) survived long enough to record 60 minutes of myth-busting, moral-panic-smashing, and one rabbit hole about blue book exams that we promise is worth it.
🔑 5 BIG TAKEAWAYS
Bad Science Is an Infinite Scroll.
A 206-page pre-print + 54 college kids + zero peer review ≠ definitive proof of anything (except how fast outlets will copy-paste an AP wire).EEG ≠ ESP.
Brainwave spikes can tell you something’s happening, not what is happening—especially if the control group’s prompt game is “we didn’t give them prompts.”Correlation vs. Causation—AGAIN.
Calculators didn’t kill math, spell-check didn’t nuke literacy, and ChatGPT won’t delete your IQ. Tools shift the work; they don’t erase the need for thought.Fear Sells; Critical Thinking Saves.
Before you retweet that “AI makes humans obsolete” headline, ask: Who wrote it, who benefits, and where’s the data? (Pro tip: “for the children” is not peer review.)Hype Your Friends, Not the Panic.
The real assignment this week? Make a PowerPoint of why your favorite human is awesome. Seriously. Humanity > hot takes.
📝 LINKS
The Conversation: MIT researchers say using ChatGPT can rot your brain. The truth is a little more complicated
The Wall Street Journal: They Were Every Student’s Worst Nightmare. Now Blue Books Are Back
Nature: The effect of ChatGPT on students’ learning performance, learning perception, and higher-order thinking: insights from a meta-analysis (across 51 experiments, this study found that using ChatGPT in class helps more than it hurts by a lot. It boosts grades the most, gives students a moderate morale bump, and also nudges their higher‑order thinking upward, with several “it‑depends” caveats, of course)
Vox: My students think it’s fine to cheat with AI. Maybe they’re onto something.
Second City: PowerPoint Meets Romance: Pitch Your Friend to a Room of Singles
The Cave Project Archives: AI in the Classroom: Tool or Cheating Device?
The Cave Project Archives: Moral Panic or Just Reality? Navigating Phones and Kids in 2024
Or…. what if this whole study is actually a study about using ChatGPT to review data and sharing AI slop without even reading it via @unoriginal_sin… “this research paper wasn’t just studying cognitive offloading, it was performing it”
Tiktok failed to load.Enable 3rd party cookies or use another browser
📊 YOUR TURN
What’s YOUR litmus test for hearing about a new study? Vote here, and we’ll round up the smartest answers in a future episode.
💌 BEFORE YOU GO
If you loved this episode, smash that subscribe button, forward this post to one headline-sharing friend, and leave us a five-star review on Spotify (it feeds the algorithmic overlords).
See you on the internet,
Jenny and Greg Swan
The Cave Project
P.S. Favorite thing on the internet this week? Jenny’s obsessed with NYC’s “Pitch & Pair” live dating decks; Greg’s busy proving he’s human with an eyeball-scanning Worldcoin orb. What’s yours?
Share this post