Prof of Ed Psych & Learning Sciences at UNC-CH | Scholar, speaker, consultant studying how people learn in the digital world | APA & AERA Fellow | Journal & Handbook Editor | Book Author | Views are my own. https://linktr.ee/jeffgreene
The problem with students not writing as much as they used to is not that there is something intrinsically good about writing but rather that writing is a great way to productively offload one’s thinking for evaluation, reflection, and elaboration. Those are the processes we cannot lose.
South Park's Eric Cartman WritingALT: South Park's Eric Cartman Writingstatic.klipy.comYet another piece of evidence supporting @karaswisher.bsky.social’s argument that you need people from a variety of backgrounds in the room discussing GenAI features. Someone in the deployment group should’ve said, “Um…can we think about what people will do this for a moment, before releasing it?”
404 MediaUPDATE: After the publication of this piece, Google sent 404 Media a statement saying it was “rolling back this feature in Google Earth while we work on implementing stronger guardrails.” www.404media.co/google-earth...
Strategy instruction enhances K-12 students' argumentative writing performance, both after instruction and in the future. Such an impressive body of evidence! Here's hoping more educators adopt the methods described here. doi.org/10.1037/edu0...
Fantastic news! Very excited to see UNC-CH building out its commitment to the Arts! www.newsobserver.com/news/local/e...
www.newsobserver.comThis is a fun one. Refutation podcasts helped participants better explain why learning styles aren't "a thing" but all podcasts (including one without refutation) decreased belief in learning styles myth. Could be a topic-specific effect? doi.org/10.1016/j.le...
"Multiple sources emphasized to WIRED that OpenAI's models also seem to have escaped containment because of lapses in implementing foundational security best practices" I mean...we all knew this where we were going to end up, right? GenAI doesn't "go rogue" - humans "make mistakes."
OpenAI’s Hacking Debacle Was a Human MistakeIf the generative AI giant had followed well-known security best practices, it’s likely that its AI agent would never have escaped to the open internet and hacked multiple companies.www.wired.comFilm you've watched more than six times with a gif. Hard mode: No Star Trek, Star Wars, LOTR, Marvel, Disney or Ghibli.
Timothy Robins in Hudsucker Proxy: You Know, For Kids!ALT: Timothy Robins in Hudsucker Proxy: You Know, For Kids!static.klipy.comAndrew JacobsFilm you've watched more than six times with a gif. Hard mode: No Star Trek, Star Wars, LOTR, Marvel, Disney or Ghibli.
Listen, in this day and age, I'm grateful for organizations that put $ behind improving education. Even if past initiatives were...less than successful. What struck me about this report's examples, however, is they don't need #GenAI. Good old-fashioned learning analytics does the job well. (1/2)
The 74The Gates Foundation announced an updated strategy for U.S. schools Wednesday, focusing on helping young people earn credentials and betting big on AI tools bit.ly/3S0RswP
The research on study strategies such as spaced practice, interleaving, and generative note-taking is pretty darn solid. That said, we need more research on how these strategies affect learning in authentic contexts, and we need more research, like this, on their combined effects.
Note-Taking Reduces the Long-Term Learning Benefit of Interleaving Over BlockingInterleaving items from different categories tends to be better for learning than blocking items by category. Surprisingly, little research has investigated whether note-taking interacts with study schedule to influence learning. Participants studied paintings from different artists, with paintings by half of those artists blocked by artist and paintings by the other half mixed with paintings by other artists. Some participants were instructed to take notes, and some were not. Then, 1 week later, participants classified new paintings by the studied artists (half the note-takers were able to use their notes on the test). Interleaving was more effective than blocking for learning in the no notes condition, but this effect was significantly smaller when participants could take notes and eliminated when they could use notes on the test. Across all conditions, participants reported that blocking had been better for learning. This study has implications for learning in educational contexts.doi.org"The breach happened when OpenAI was testing one its latest AI models against ExploitGym, a benchmarking framework that scores AI systems on their ability to find and exploit software vulnerabilities..." (1/x) www.wired.com/story/openai...
OpenAI’s Rogue AI Agent Hacked More Than Just Hugging Facewww.wired.comWhat I found most interesting here was the discussion of the potential limits of neural network GenAI and the (partial) bet Google may be taking on world models being the true path to artificial general intelligence, instead. www.thealgorithmicbridge.com/p/the-actual...
The Actual Reason Why Google “Fell Out” of the AI Race Changes EverythingHottest story of the yearwww.thealgorithmicbridge.comIntriguing discussion of symbolic representations of higher-order thinking v. purely neural/connectionist ones. infinitefaculty.substack.com/p/symbols-ne...
Symbols, neural networks, and mathematical intelligenceSome reflections on connectionism and the basis of higher-level cognitioninfinitefaculty.substack.comFull disclosure: I may be out of my depth here, but I read this article as arguing for metacognitive (what they call "metalearning") planning processes including not just forward search but also precomputation and refinement of models of the world. Sound like metacognitive knowledge & skills, to me.
Educators have to re-examine their learning outcomes, separating the verifiable and mechanical tasks from the non-verifiable, human ones. The former can be taught with GenAI, the latter must not. Expertise comes from the desirable difficulties and lessons learned from human work.
If You’re Over 40, You’re Ready to Use A.I.Opinion Article in The New York Times by Debbie Millman titled "If You’re Over 40, You’re Ready to Use A.I."www.nytimes.comPeople I respect have said #GenAI cannot do nonpositivist qualitative research, full stop. I do wonder if GenAI can play roles in such work, however. This article describes a way to think about those roles, focusing on GenAI as augmenting the human work of data collection & thematic interpretation.
Towards a methodologically congruent framework for GenAI use in nonpositivist qualitative researchGenAI is increasingly used in qualitative research, yet much current practice remains methodologically incongruent. Attempts to make LLMs replicate qualitative coding often create a mismatch betwee...doi.orgMore evidence to support my sense that the "intelligence" in various models (e.g., ChatGPT, Claude, Kimi) is rapidly becoming a commodity. What will matter is the "harness" (i.e., the app interface, connections to other services, etc.) and the cost - those will be the differentiators among models.
Six-Chart Sunday – Shooting the Moon: What the New Chinese AI Model Means for Policy & Markets6 Infographics + 1 Video (30 years ago this week… Strug nails the landing)brucemehlman.substack.com