Infinite Patience Is Not Good for Education
How many times does Sal Khan get to fail?
One of the favorite selling points of tech industry figures when it comes to the intersection of AI and education is the capacity for the creation of an on-demand tutor that is “infinitely patient.”
This is Sal Khan’s promise in Brave New Words: How AI Will Revolutionize Education (and Why That’s a Good Thing), where he takes inspiration from the novel Ender’s Game (a famously positive story about the fates of children) and describes a world where students have “a personalized tutor in every subject.”
In his manifesto, “Why AI Will Save the World”, tech investor Marc Andreessen puts it this way:
Every child will have an AI tutor that is infinitely patient, infinitely compassionate, infinitely knowledgeable, infinitely helpful. The AI tutor will be by each child’s side every step of their development, helping them maximize their potential with the machine version of infinite love.
In my review of Khan’s book I called his views “unserious” because he does not seem to understand much of anything about teaching and learning, skipping several steps on his way to his AI-Aristotle future. This idea that somehow infinite patience is a positive quality in a teacher is perhaps the most telling sign that these people simply do not know how learning works.
Some of my most important formative educational experiences involved some teacher or authority figure losing patience with me. I will never forget the moment that my 8th grade language arts teacher Mrs. Thompson told me to “cut the shit” when I had quarter-assed a writing task. This is the same teacher who also read a short story of mine and declared that “she’d never read anything like it.” I think this was praise, or at least I took it that way, so when Mrs. Thompson said the obvious about my lackluster effort, I felt bad. I’d disappointed her.
Humans learn in communion with other humans. Maybe this was not true for Sal Khan who is quite obviously smarter and more driven than most other humans, but it is true for the vast majority of us.
In a recent article at Chalkbeat Sal Khan admits, with some mix of surprise and regret, that the very technology he said was going to revolutionize education was a “non-event” for most students. “They just didn’t use it much.”
No kidding. Back when Sal Khan was publishing Brave New Words, lots of people were saying this was going to be the case, me, Dan Meyer, Audrey Watters, and many many others. As Watters say, “AI tutors" are not the future; they're the past.”
Please enjoy this mashup of a BBC report from 1965 on “learning machines changing education” and Khan’s demonstration of what it’s like to work with the Khanmigo chatbot.
In the Chalkbeat post mortem, Khan Academy’s chief learning officer Kristen DiCerbo suggests that the Khanmigo failure is a user problem, “Students aren’t great at asking questions well.”
My experience is that students are great at asking questions, which is not the same thing as asking questions of a tutor bot.
Back in 2024 I “debated” some dude from the American Enterprise Institute on the potential of chatbot tutors like Khanmigo. I took the skeptic’s side and argued that the number one reason this technology would not revolutionize education is because it did nothing in terms of the chief challenge of education: engagement.
I was right right, AEI dude was wrong. Sal Khan was wrong. Bill Gates, Laurene Powell Jobs, Arne Duncan, Adam Grant, Angela Duckworth, Tony Blair, and Francis Ford Coppola (seriously) - all of whom blurbed Brave New Words declaring it very serious indeed - were wrong.
They have always been wrong. They were wrong before they got started and yet hundreds of millions of dollars have gone towards a project that was doomed from the outset, dollars that could have - at least in theory - gone to, I don’t know, human beings who teach.
Over the years I have tried to practice good faith to these projects even as I was skeptical. I have spoken at length to people inside these projects and the granting organizations that fund them and it is clear they are well-meaning, not “grifters” out to profit at the expense of others.
But, like Mrs. Thompson telling me to cut the shit, I am out of patience. It’s time to move on from people like Sal Khan and projects like Khanmigo from sucking up so much money and oxygen when it comes to our systems of education.
Sadly, this is not happening. Not unless we make it happen.
We’re up against billionaires, world leaders, thought leaders, politicians, and one of the greatest film directors of all time, a tough set of opponents, but at least we have the benefit of being correct.
What do you do when your AI-schools revolution fizzles? Slink off to lick your wounds and figure out something else to do with your life? Go back to the drawing board at a root level to better understand where you went wrong and return with something more deeply considered?
Not Sal Khan. You pivot to proposing a “disruption” of higher education.
Welcome to the Khan TED Institute, a joint project between Khan Academy, TED (as in talks) and ETS (Educational Testing Service). For $10,000 they are proposing to create a bachelor’s degree in “Applied AI” delivered online in two years.
I aired my doubts and grievances over this project at Inside Higher Ed in which I declare this project what it is: bullshit.
But it is bullshit backed not just by these three wealthy and powerful nonprofits, but also the institute’s “corporate thought partners” including Google, Microsoft, McKinsey, and Bain (among others).
Lest my feelings about this project be unclear, this is how I put them in the column:
These people are my enemies. I have only ill will for this project and wish them failure, because this vision for a future of postsecondary education is a recipe for mass immiseration and public disempowerment. Imagine a world where Microsoft, Google, McKinsey, et al … get to determine what and how you learn from cradle to retirement.
Anyone involved in higher education, particularly public higher education or private higher ed where your institution is not insulated by wealth and privilege, should also view this project as a direct assault on their continued existence. The higher education sector and those who have historically been responsible for it (government, voters, etc. …) should pause and reflect on how what’s happened to the sector has made it potentially vulnerable to this sort of program, but we also must set recriminations aside and deal with the threat directly.
Again, I don’t think Sal Khan is a bad person. He is not evil. I wish him nothing but health and happiness in his day-to-day existence. But we should wish him nothing but ill-will on this project because its success will mean that we have a society where individuals are essentially indentured to these corporate thought leaders. It means distorting education into a shape that pleases tech industry giants and consultants. To the extent our society already operates this way is not to our credit.
That a group of people think this is a positive direction for our collective future is mind boggling.
The good news is that Sal Khan has failed in every one of his revolutions. He gives good TED Talk, but he doesn’t seem to understand or care for anything about pedagogy. As Dan Meyer says in his obituary for chatbot tutors, “Given that Sal Khan has tried unsuccessfully for nearly two decades to abstract humans away from human systems—first with human explanation, then with human evaluation, and most recently with human tutoring—it seems unlikely that he is the right person now to pivot edtech towards humanity.”
Still, I wouldn’t mind seeing the Khan TED Institute ethered from the get-go rather than having to engage in the same I-told-you-sos a couple of years from now.
Khan made it to 60 Minutes twice, 12 years apart, each time talking about a revolution. He was wrong both times.
How many times does someone get to be wrong before we stop listening to them?
Why do we have infinite patience for this man?
Links
At the Chicago Tribune, I tried the LitRPG genre and while I think I get what others get, I don’t get it.
At Academic Freedom on the Line I contrasted the end of Hampshire College with the birth of the Khan TED Institute and what that says about the world today.
At the New York Times, Colson Whitehead entertainingly puts the boot to using AI to write for you. (Gift link)
A doctor who was an early adopter of AI scribes stopped using them when he realized how it was distorting his practice.
The 2026 Guggenheim Fellows were announced, including lots of writers. I always get a little envious of these things and then I remember that you actually have to apply, which in my case is never going to happen.
Via my friends McSweeney's, “In Our Glorious AI Future, There Will Be No Such Thing as Money (For You)” by Andrew Singleton.
Recommendations
1. Fintech Dystopia by Hillary J. Allen
2. The Road: Stories, Journalism, Essays by Vasily Grossman
3. Day of the Oprichnik by Vladimir Sorokin
4. The Mountain in the Sea by Ray Nayler
5. Carthage: A New History by Eve MacDonald
Sean H. - New York City
A bit of a tough one for me since I don’t know these books, so I’m falling back on my biblioracling gift and letting the spirits guide me: The Thousand Autumns of Jacob de Zoet by David Mitchell.
I’ve got a bit of a backlog, but I also have a combo work/pleasure trip next week so I may do an all recommendations newsletter to catch-up so don’t hesitate to ask.
I’ll be at a conference of language arts teachers for the province of Alberta next week in Banff, which is very exciting on multiple fronts.
If anyone has Banff-related travel tips, please share them in the comments.
See you, in some form, next week.
JW
The Biblioracle
Blaming students for being bad at asking questions shows an incredible lack of understanding about student curiosity and what LLMs are capable of. It's almost like these folks have never been in a classroom!
LLMs are good - except when they're making things up - at pulling up facts and summarizing whatever relevant text they can find. Unsurprisingly, that's generally not the kind of information that students are likely to be more curious about, as they are also capable of searching the internet for information.
We recently wrapped up a unit about the Civil Rights Movement in one of my history classes, and students had all kinds of questions, including why the reaction by white southerners to school integration was so over the top. They've grown up in a relatively tolerant part of the country in an era in which, up until recently, being a racist asshole was verboten. It's not a question of fact that can be answered by a chatbot. It's a question of human beliefs and motivations and what has - and hasn't - changed over time.
I really appreciate this perspective! Over the last few years, I’ve had some very challenging students. Love lost my patience many, many times. And each time, worry that I’ve damaged our relationship or discredited myself as their teacher. I keep coming back to the imperfect messiness of teaching and learning. The students who I have lost my patience with the most, those are the same students who come back to visit me. Why? Because I was an imperfect human teacher who cared, even when I lost my patience. In fact, that’s why I lost my patience (and why Mrs. Thompson lost hers with you). I cared because I saw more from them than what they were giving.