- On EdTech Newsletter
- Posts
- Interesting Reads This Week
Interesting Reads This Week
Back to regular programming with a mini-rant about AI
Glenda Morgan
July 06, 2024
Was this forwarded to you by a friend? Sign up, and get your own copy of the news that matters sent to your inbox every week. Sign up for the On EdTech newsletter. Interested in additional analysis? Try with our 30-day free trial and Upgrade to the On EdTech+ newsletter.
Last week's uncharacteristic burst of optimism has given way to more typical (for the past two years) EdTech-related despair. This occurred even despite my Fourth of July triumph (and sugar high) of making the best banana pudding ever, homemade vanilla wafers included! What did me in this week?
The hype-iest AI hype that ever hyped
This week, I seemed unable to avoid endless stories about how AI would reinvent learning, becoming the solution to all our problems and the answer to all our prayers. On the university side, I saw too many stories like this one from Morehouse about combining generative AI and XR to create spatial classroom assistants who can answer student questions.
Similarly, Chegg (despite the change of leadership) seems intent on doubling down on the message of using AI to personalize student learning. The new CEO made some interesting arguments in a recent interview.
Students could have a learning assistant in their pocket that anticipates their needs and adapts to their strengths and weaknesses. GenAI tools for learning, when fine-tuned and evaluated rigorously to align with education outcomes, can respond instantaneously and in a more natural, back-and-forth conversational interface with students. [snip]
We anticipate the platform will also let students know if they are learning at optimal times in their day, suggest students engage in more practice rather than relying only on review, and encourage students to push through difficult concepts to build learning resilience. [snip]
That’s why Chegg is launching our proprietary AI platform, including 26 subject-specific LLMs, all trained on Chegg’s 100 million plus pieces of learning content, created by over 150,000 Chegg subject-matter experts, designed with learning and accuracy in mind, and rooted in learning science.
Even apart from questions about Chegg as a tutoring service, the kinds of assumptions implicit and explicit in these claims are starting to drive me crazy. Because a technology has a set of affordances, how it is used makes a big difference. The fact that an instructor created some content does not necessarily mean that it is rooted in learning science. Pedagogy matters.