Meet The New Face of Traditional Education
How Coursera Is Becoming What It Tried to Reinvent
I see it all the time.
A brand new, innovative conference is announced that will upend traditional education and usher in the age of collaborative learning. But the program is a list of traditional 45-minute lectures, followed by 10-15 minute Q&As.
Or I get excited for an online course that advertises itself as truly collaborative, “ground-breaking” (eww), and “game-changing” (double eww). But the course is nothing but a series of short video lectures and brief quizzes.
Hardly innovative.
In these instances, there is a wide gap between the content and the form in which that content is delivered. A person touts new age technology’s ability to connect people and to challenge traditional learning. But then they tell us to sit down, shut up, and listen to their 45-minute presentation with mind-numbing PowerPoint slides.
I think platforms like Coursera are falling into the same trap.
About a week ago, Coursera announced a new set of AI-powered features. Below are a few paragraphs from the company’s CEO, Jeff Maggioncalda:
Here is the TL;DR with my own snarky commentary:
AI-Assisted Grading: Apparently, this is being listed as an Academic Integrity measure. I’m not sure how or why. It feels like this was snuck in.
Proctoring and Lockdown Browsers: We’ve had them for a long time, and I’m not sure if this even counts as AI (depending on how they do it).
Viva Exams: These appear to be similar to graded ChatGPT chats, where students dialogue with a chatbot for a grade.
Dare I say, these features feel…well…like traditional college education.
I’m talking about written exams completed under duress, with the professor and Teaching Assistants walking around the room and peering at anyone who dares to look off to the side.
I’m talking about putting weeks of work into a paper addressed specifically to your professor, to have it graded and commented on by someone else.
A Failed Attempt at AI Colleges?
I wrote this a long time ago on LinkedIn.
That was my prediction. AI colleges would emerge as significantly cheaper options. The worst ones (which probably wouldn’t last) would run almost completely with AI avatars and professors. The better ones (which I thought would have a better chance of upending the traditional college) would balance human labor with machine labor.
The better versions wouldn’t be quite as cheap. But my guess was that they’d be longer lasting.
also saw something similar coming. He wrote about it in his own Substack, The Future of Higher Education.My argument: this recent set of Coursera announcements gives us a glimpse into what one version of the AI college will look like.
It’s just not a particularly good version.
Leon Furze, an educator and AI consultant, identifies the problem succinctly in his own LinkedIn post:
He’s right.
The model here isn’t an innovative, reinvented classroom.
It’s a doubling-down on the Industrial Era style of education, using AI as its enabling mechanism.
In this line of thought, AI isn’t a chance to rethink how we engage with our students. It’s a way to surveil and control.
Beneath the Hood (?)
After learning about Coursera’s updates, I started thinking.
A lot of this depends on being able to detect AI work. So, I took to the internet!
Lo and behold. This was published in Inside Higher Ed on June 11th, 2024.
So, a quick recap. On June 11th, Coursera writes a post announcing a list of new AI-powered Academic Integrity updates. On June 12th, news breaks that Coursera has launched its own AI Plagiarism Detector.
Now it makes sense.
Coursera needs something under the hood, to make it all work.
They need their own Detector, because then they don't need to rethink. They can scan assessment to (supposedly) distinguish between AI work and human work.
As long as we trust the AI Detector, we don't really need to innovate.
So, what's the problem.
For one, AI Detectors don't work. This isn’t news.
has covered this consistently over the last year. This is one of his breakdowns.I’ll also add: even if they did work, relying on them gets us into a cat-and-mouse game that we’ll likely lose in the long run.
We can either double down on those detectors (as Coursera seems to be doing) or we can find other, more sustainable ways of moving forward. I choose the latter.
Let’s Rethink Instead of Reinforce
Coursera is reinforcing traditional education instead of rethinking it.
We can do better.
In fact, there are some dedicated, hard working people who are already doing better.
Here’s a list of some of them.
- : Follow his AI EduPathways newsletter.
Amanda Bickerstaff: She’s been helping many schools and district adapt.
Leon Furze: He’s written about basically everything.
Merissa Sadler-Holder: She left teaching to specifically try to figure this out.
- : He just wrote a book about just this very topic.
Sabba Quidwai: She’s been doing some great work on redesigning schools.
- : The man has built schools. He’s now rebuilding them for The Age of AI.
- : He’s rethinking modern education around content mastery.
- : He’s thinking a lot about AI and the future of writing.
- : His Educating AI newsletter is fantastic.
Likely, you won’t agree with everything they have to say. And that’s the point.
If we do enough digging into any of these voices, we notice a level of nuance that is sorely lacking in conversations about AI and Education.
That’s exactly what we need.
"In these instances, there is a wide gap between the content and the form in which that content is delivered. A person touts new age technology’s ability to connect people and to challenge traditional learning. But then they tell us to sit down, shut up, and listen to their 45-minute presentation with mind-numbing PowerPoint slides". This is one of the points I appreciated most in the issue. Honestly, I don't write and I don't have a background related to education, but this perception is very widespread and I perceived it very directly. Furthermore, there is a crucial point that in my opinion should be highlighted in what has been stated: "sometimes, instead of reinforcing, we should rethink". We often try to integrate new and old dynamics without thinking about the overall vision. This integration that you underlined, however, is an important process even if only to stimulate the rethinking of certain dynamics, especially in education.
I feel like the real attraction of this AI courseware has less to do with the students than the faculty. The more administrations can AI-ify the less they need professors, who are already outnumbered 2-1 in many institutions by administrators.