🥊 Education vs. the big AI companies

🔍 A better alternative to AI detectors

AI detection software is big business in education.

Why? Because there’s demand.

Educators — and education leaders — are clamoring for it. They think it’s a solution to what they view as the problem: students turning to AI to do their classwork for them.

(I mean, that is a problem … but there are so many other competing problems that don’t involve this, but that’s a post for another day …)

The real problem? AI detectors are wildly inaccurate. Even the best ones only predict AI use accurately 80 percent of the time.

Adam Sparks, an English/social studies teacher turned edtech developer (shortanswer.com), smelled a rat. 🐀🐀

In his guest post on my Ditch That Textbook blog, he wrote …

TurnItIn, the largest and most influential of the AI detection companies, recently published a whitepaper in which they claim to show that, “Turnitin’s AI writing detection system has been independently shown to have high effectiveness in correctly identifying AI-generated content.” 

When I dug into this claim, I found it was partly supported by the following research: “Weber-Wulff (2023) showed that Turnitin’s AI writing detection system outperformed all other AI writing detection solutions on the market in accurately detecting AI writing across a diverse set of evaluations.”

The Weber-Wulff study above showing that TurnItIn’s AI detection software was wrong over 20% of the time? The same one that concludes “these systems should not be used in academic settings”? Yet TurnItIn cited that study as supporting their product.

One conclusion I come to here is that TurnItIn assumed that no one would read the research they cited.

Another conclusion I come to is that TurnItIn is knowingly lying, by omission, to teachers.

Long story short:

• Adam called out TurnItIn on social media and got LOTS of attention.

• TurnItIn’s VP of AI met virtually with Adam. The meeting went well, where he said (among other things): “You should never give a kid a zero based on TurnItIn results alone” AND that TurnItIn should instead be used as a “healthy conversation starter”

What can we learn from all of this? Here are some things I’ve been thinking about …

First, instead of adapting to a changing world AND adapting ineffective teaching practices, some educators are trying to protect the status quo with some traditional teaching practices that never really worked well in the first place. And they’re using AI detectors to dig their heels in to maintain that status quo.

Second, many educators don’t know the realities of these AI detectors. Why? For one, companies like TurnItIn aren’t truthful about their AI detectors … and educators just believe what they’re told.

Third, educators aren’t getting training and professional development about AI and its use in education. (PS: I can help with that.)

Bottom line: It’s time to have the talk with teachers.

The talk about AI detectors and their ineffectiveness.

It also might be time to talk about teaching practices … to evaluate whether they’re getting the results that we want and if they’re really preparing students for the future they’ll experience.

In the post, Adam offers several suggestions for a path forward that doesn’t rely on AI detectors.

Read all of Adam’s guest post here: AI detection software doesn’t work. Do this instead

🗣 DISCUSSION BOARD: Our changing views on AI

Last week’s question: How have your views changed on AI and its implications on education over the last year? (Or how have they stayed the same?)

👀 View the whole discussion here. (It’s not too late to participate! If you haven’t, join our community to reply. It’s free and quick.)

Some highlights and summarized responses:

• “My view has always been to allow AI to be a thought partner for teachers and take advantage of it to save time and let teachers focus more on relationships with students instead of tedious work/tasks. I truly believe that AI can help teachers in so many ways, and my views have just been solidified over the past year.” — Brent Mischnick

• “I have learned how much it can be used to support the work of educators, to save them time, to elevate the opportunities for different lessons and ideas, and enhance learning. Also scratching the surface on how helpful it can for students with special needs or differing learning needs. In creating our guidelines we in curriculum, worked alongside our special education team as well as privacy and IT. It was a good learning experience and the documents created were robust. As soon as they are finished and vetted, I will share them here.” — Lorrie Temple

• “I do believe more in AI because it saves time to find the appropriate information etc......On the other hand I always feel that I need to put my input to make it more practical for my students. As a teacher, I know the right level, expression, word for my students.” — Daniel Bartolo

• “Changed a lot. But I found that educating myself has shifted my thoughts on AI as a negative thing, to something more positive. More practice and more research on my part will help.” — Ken Wong

Also, if you didn’t catch it, Chad Sussex shared his district’s school board-approved policy on AI in the educational environment.

He says they’re working on an addition around nonconsensual image/video … something I haven’t seen much discussion about. Thanks for sharing this Chad!

📺 I talked about AI on the Dr. Phil show

If you haven’t heard, I was a guest on a new episode of the Dr. Phil show! Topic: AI in Education.

The whole experience was … well … quite an experience!

📺 Watch it for free online here. 

I was joined on stage by Jean Darnell, a terrific school librarian from Austin, Texas. We represented what I’d like to call the “common sense middle ground.”

The whole first half of the show was a big production about a private school called “The Alpha School” that teaches students for two hours a day on AI chatbots. The rest of the day is what they call a “life skills workshop.” Oh, and the school doesn’t employ human teachers … just non-licensed adult “guides.”

The second half covered topics in education related to AI. I talked about the importance of the human teacher, the existence of bias in AI models, and how we need to strike a balance between human and AI in education.

👀 READ MORE: I have a big section of my Ditch That Textbook email newsletter tomorrow about it. (Subscribe for free here so you don’t miss it.)

🗓 This week’s discussion topic

Here’s this week’s new discussion topic (posted under AI Professional Development):

FYI: Hit reply to this email if you want to suggest a future discussion topic!

Generative AI can create products. Often, we grade students based on the product (research papers, essays, answers to questions, math problems, etc.). In my presentations, I'll often say, "When AI cheapens the product, focus on the process."

✍️ Click here to view the AI Professional Development discussion board to reply

🎙 Quote of the week

AI detection leads us down the wrong path in adjusting instruction and assessment post-AI. It represents the easy-button option - “I don’t have to change my instruction and assessment, I’ll just tell kids not to use AI and use detection software”.

— Adam Sparks, educator and creator of Short Answer (see post)

📚 New AI resources this week

1️⃣ AI first steps for teachers (via Digital Learning Podcast): In the latest episode of the Digital Learning Podcast, Matt and Holly share some steps educators can take to get up to speed about AI.

2️⃣ How to Create a Responsible Use Policy for Your District (via Merlyn): In the first of a multi-part series, Merlyn suggests some steps that districts can take.

3️⃣ Students And Teachers Of Color Are Embracing AI In Schools At Greater Rates Than Others. Why? (via Forbes)