The Alpha School (originally based in Austin, Texas) came to my attention recently because of their use of AI in education. The school makes some ambitious claims: depending on the school location, they claim their students test in the top 1% or 10% of students on the MAP test, and that they complete their academic curriculum in focused sessions of only 2-3 hours per day. This leaves them open to spend the rest of the day working on what appears to be project-based learning of a more career-focused nature. Moreover, they say these intense study sessions are enabled by an AI-powered learning platform.

Now, how seriously should we take their claims? I won’t dispute the fact that their students perform at the levels they report, though it does seem germane to point out that a school with a tuition of $40,000 a year is likely to attract students that would perform well academically regardless of the way the curriculum was implemented. On the other hand, as Sal Khan pointed out in his TED Talk on the potential of AI tutors, fairly solid research backs up the idea that tutoring is proven to be the most effective mode of instruction, and can lead to gains of up to two standard deviations on academic performance.

By the way, there’s been some pretty silly responses to Sal Khan’s talk. On one podcast, a guest weighed in on the value of tutors:

So let’s just talk about this. It’s actually not as complicated as they make it seem in all of these TED Talks and all of this stuff. Teaching to me, wasn’t that complicated. It’s like, here’s what I want you to know. Here’s how I get you in the mood for learning. And then we’re going to just discuss it until I know that you know what’s going on. I didn’t even have very many tests in my classroom because I could tell which kids knew what they were talking about and which kids didn’t. And then the ones that didn’t, you just hold them back for a second. Come see me at lunch, the three of you, and let’s discuss this. And why didn’t you get it? And what happened? And many times, it’s not that they didn’t want to do it. It’s that they have a ton of other stuff going on.

I’m not sure how long she was a teacher, but I haven’t met many veteran teachers who have described teaching as “not as complicated as they make it seem.” Granted, I don’t know exaclty what she taught; an approach for getting kids ready in ELA might look a little different than, say, advanced calculus. In terms of international competition, I don’t think the United States is as concerned with making a gain of two standard deviations in writing and reading as we are in making sure we have an adequate supply of skilled students for the STEM professions. Despite the pooh-poohing of this podcast, it really does seem as though personal tutoring delivered via AI could have the potential to help push more students into that skilled range.

If we grant that Alpha School has cracked this code and produced something capable of delievering those ideal two standard deviation results, what are we to say about that $40,000 price tag? Is it essentially a paywall keeping middle- and lower-class students out of huge potential gains in achievement?

If so, it seems it would be a great boon to the common good for people to start work on reverse engineering this curriculum to create an open source equivalent. Even groups like Khan Academy could be doing more to make this happen. KA does license its videos as Creative Commons, and they support a lot of different open source projects, but a lot of them have fallen by the wayside, including the Khan API. A lot more could be done to make their AI Khanmigo open source as well.

What we need, ultimately, is not more groups trying to capture attention and prestige for themselves, but public-minded organizations willing to donate their time and efforts to the greater good.