WN #11

Why This Time is (Actually) Different

One of the biggest hurdles facing the adoption of generative AI by educators is that it often gets lumped into the greater “EdTech” bucket, which doesn’t always have a great reputation. Sure, there has been plenty of helpful technology that teachers have integrated into their routines over the years, but if you are a teacher who has spent any time in the system, chances are you’ve encountered plenty of new edtech pitches, platforms, and tools that were supposed to revolutionize your classroom but never delivered on the hype, eventually finding their way to an ever-growing edtech scrap heap. And while generative AI fits the definition of educational technology, and makes some of the same promises we’ve heard from edtech products before (e.g. personalizing the learning experience), we are actually entering a new technological frontier that we can’t afford to ignore. Here are some reasons why generative AI tools should be viewed differently from traditional edtech offerings:

Accessibility: While there are plenty of new products that incorporate generative AI and would love to sign you up for a subscription plan, access to generative AI technology is freely available for any student, educator, school, or district to use, so long as they have an internet connection. It’s as simple as visiting a website or downloading an app, and requires no additional skills to get started other than being able to converse in plain language (whether that’s through typing or speaking).

Ubiquity: With accessibility comes ubiquity. Generative AI is already starting to show up in more and more aspects of our lives, from social media to the doctor’s office. Students and educators alike will be interacting with the technology with increasing frequency whether they are ready for it or not. And while most students engage with traditional edtech because their teacher tells them to and it’s part of the class, generative AI presents a fantastic opportunity for authentic learning and exploration that both students and educators can embark on together, providing crucial knowledge and skills that extend well beyond the classroom.

Customizability: This isn’t a product that has been designed to tackle a specific problem in a specific way (and, as with traditional edtech, often designed by people who are far removed from any real-life classroom). Generative AI allows anyone to build, create, and use according to their needs and desires. What works great for one teacher, school, or set of students might not work at all for another. This is one of the reasons why we have such a patchwork of educational technology in our schools in the first place. With generative AI, I can use it to quickly create differentiated assessments, curate a playlist of songs centered around a specific theme, and then design my holiday cards not because I was told that’s what it does, but because I told it that’s what I want it to do.

Nascence: This isn’t a finished product being rolled out with an occasional update. We are still in the early stages of understanding and harnessing the full capabilities of generative AI — a period of experimentation, learning, and adaptation. This gives educators (and students!) a unique opportunity to shape its growth and direction, ensuring that generative AI complements rather than complicates the learning process, and provides students with the tools they need to succeed in the world they will be living in, rather than confining them to the blueprints of the past.

GenAI and SEL

While most of the talk around generative AI and education has centered on academic integrity, a recent survey by the Center for Democracy and Technology found that, of students who used generative AI during the 2022-23 school year, just 23% said they used it for academic purposes, whereas 29% said they used it to help deal with anxiety or mental health issues, 22% for issues with friends, and 16% for family conflicts. This data suggests we should be placing more focus on the impact generative AI may have on social-emotional learning, for better or for worse.

AI Skillsets for the New Era

As the AI landscape continues to evolve, educators will need to develop AI literacy on the road to becoming AI-empowered. Here are 10 AI skillsets that will help us get there.

Ideas! Ideas! And More Ideas!

If you are looking for ideas and inspiration for incorporating AI into your classroom, here are 9 Tips for Using AI for Learning (and Fun!), a list of Roles for AI to Play in the Classroom, and a document from Canva on how educators can generate and use AI images.

📌 Chat with GPT

Voice chat is now available to all free users of the ChatGPT mobile app. Previously available only to ChatGPT Plus subscribers, the speech-to-text and text-to-speech technology enables users to have a verbal conversation with ChatGPT that is also transcribed and saved if you want to come back to it later.

📌 Claude Gets an Upgrade

Claude, one of ChatGPT’s biggest competitors, got a recent upgrade that gives it a larger context window (it can now process up to roughly 150,000 words or 500+ pages), and decreases its hallucination rate (it is less likely to give you incorrect information). Other key aspects of the upgrade mostly impact developers for now, so I won’t dig into them here. Claude is a great (free) option if you want to upload files as part of your conversation.

📌 Bard Can Search YouTube

Google’s Bard chatbot can now answer questions about the content of YouTube videos. Bard has lagged behind competitors in terms of quality, but adding niche features like this one may give more people a reason to use it, especially if it can replace the need to scroll through videos to find specific information.

Generating Sentence Frames

I usually spend a decent chunk of time writing sentence frames to help students at various levels develop their writing skills and academic vocabulary. Using AI to help with this process can be a quick and easy time-saver. Here's an example using ChatGPT:

A Chatbot that Cites its Sources

Looking for an AI chatbot that cites its sources? Check out Perplexity AI, which is free to use (and doesn’t require any sort of sign up). Bing Chat will cite sources as well, but I often prefer Perplexity’s responses. You can also ask Perplexity to focus its search to sources from the web, academic papers, YouTube, Reddit, or Wolfram Alpha.

• You may have noticed that Apple has been noticeably missing from the AI frenzy, they’re working on it. Amazon is developing its own AI model as well.

• YouTube is experimenting with AI tools that allow users to create their own music tracks incorporating artists’ voices and instruments simply by typing a prompt or humming a tune. Google has its own text-to-music generator experiment.

• Why visit the doctor when you can just pop into an AI-powered CarePod?

• Text-to-video and text-based image editing are coming to Facebook and Instagram.

That’s all for this week! If you appreciate the content in this newsletter, consider subscribing for free, or sharing with people in your network who may find value in it. If you are looking for more, feel free to check out the archive to catch up on any editions you may have missed.