GM. This is The Great Unlock, your weekly cheat sheet for the AI revolution. We filter out the noise, red pen fake news, and give out stickers to those deserving.

Let’s unlock it.

The social media era is ending. Not because people are using their phones less, but because AI is dismantling the structural conditions that made social media dominant in the first place. The habit loop is collapsing. The shared feed is fragmenting. And the internet-era platforms built on those foundations are discovering what Clayton Christensen called the Innovator's Dilemma: past success is becoming the very obstacle to future relevance.

K-12 education is on the same trajectory. Most of the EdTech tools in your district today were built on internet-era logic—broadcast content at scale, track completion, optimize engagement. AI will not make those tools faster or smarter. It will make them obsolete.

The critical question is not whether this transition will happen. It will. The question is who will shape what comes next: the algorithm, the platform, or the school district.

The internet did not produce faster newspapers. It eliminated the business model of newspapers and built new economies on the rubble. Social media did not produce faster cable television. It produced TikTok—an entirely different market economy with different sources of influence, different participatory dynamics, and different cost structures.

AI is a general purpose technology of the same order. It will not produce better LMS platforms. It will produce environments in which the LMS paradigm no longer makes sense.

The futurist Sinéad Bovell has documented this shift extensively in her analysis of the social media era's decline. Her core observation—that general purpose technologies break open the present rather than accelerate it—applies with precision to the situation facing K-12 EdTech today. 

Social media's power was structural. It owned the habit of reflexive phone-unlocking. Every time attention wandered, the feed was there.

AI agents are dismantling this. The more a person can simply speak a request and have it fulfilled—summarize this, book that, flag the important thing—the less they drift toward a feed to fill cognitive dead time. The habit loop that built trillion-dollar platforms is being interrupted.

The same dynamic is coming for EdTech. The 'open the app, watch the video, take the quiz' loop is the educational equivalent of the scroll feed. When students can instead query an AI agent that scaffolds their thinking in real time, the static content-delivery model does not become less effective. It becomes irrelevant.

Of all the risks in the AI transition, the most serious is the one least discussed in EdTech vendor conversations: the collapse of the shared curriculum.

Social media's defining failure was not that it was engaging. It was that full personalization—every user inhabiting their own algorithmically curated reality—destroyed the common ground required for civic life. When every person sees a different feed, the shared references that make collective reasoning possible disappear. The consequences have been well-documented.

Full AI personalization in education poses an equivalent risk at the instructional level. If every student follows a completely differentiated learning path—different texts, different examples, different conceptual sequences—who decides what the common intellectual inheritance looks like? What shared vocabulary holds a classroom, a grade level, a district together?

This is not a hypothetical concern. Mark Zuckerberg's announcement that Meta would begin generating personalized AI content for each individual user represents the logical endpoint of this trajectory applied to media. The EdTech equivalent is already being pitched to districts: AI-native platforms promising fully adaptive, fully personalized learning journeys with minimal teacher involvement.

The districts that adopt these tools wholesale will generate impressive short-term engagement data. They will also be running an uncontrolled experiment on their students' ability to function in shared intellectual spaces—classrooms, assessments, civic life—that require common reference points.

The incumbent EdTech platforms—Google Classroom, Canvas, Clever, and the major adaptive content providers—have adopted to AI. They have announced AI assistants, AI grading tools, AI lesson planners. They market these features aggressively to districts.

But their business models, data architectures, and district relationships are all optimized for the internet-era paradigm. They cannot make the structural move required by the AI era without destroying the revenue base they were built to protect. This is precisely what Christensen meant by the Innovator's Dilemma.

A district that waits for incumbent platforms to deliver the AI-era solution is making the same bet that newspaper publishers made when they waited for their print operations to fund a digital transition. The structural incentives do not allow for it.

The most interesting counter-signal to the AI-driven collapse of passive consumption is the rise of live-streamed experiences. On YouTube Live, Twitch, and TikTok Live, participation is growing—precisely because AI makes everything else infinitely abundant. When content is available on demand, at any time, personalized to your preferences, the thing that becomes scarce is simultaneity. You were either in the live moment or you were not.

The classroom has always been the original live experience in education. It is participatory in a way no asynchronous content can replicate. Students react to each other. A teacher reads the room and adjusts. A discussion produces a thought that no individual would have arrived at alone.

As AI makes content delivery cheap and scalable, the synchronous human moment in education does not become less valuable. It becomes more valuable—provided it is used that way.

This has direct implications for how districts should evaluate EdTech. The tools worth investing in are those that make the live classroom moment more effective—not tools that substitute for it. The question to ask of every product is not 'can this scale?' but 'does this make what happens in the room better?'

HierLearning is an EdTech incubator built on a simple thesis: measurable student outcomes are not a feature. They are the only thing that matters.

Everything we evaluate, fund, and develop is organized around that thesis. We do not back tools because they are AI-native. We do not back tools because they have beautiful UX or impressive engagement metrics. We back tools that can demonstrate—with pre/post data, with district case studies, with honest accountability—that students learned more.

The AI transition creates an urgent need for exactly this kind of discipline. As AI makes it trivially easy to build EdTech tools that generate impressive demo experiences, the gap between tools that look good and tools that produce outcomes will widen. Districts without a rigorous evaluation framework will be sold novelty and call it progress.

The criteria HierLearning applies to every tool it evaluates, funds, or recommends:

•       Measurable learning outcomes. Not engagement. Not completion. Not teacher satisfaction scores. Evidence that students know more, can do more, or retain more than they would have otherwise.

•       Personalized pathways, shared destinations. Any personalization engine must be able to answer: toward what common outcome? And who audits it?

•       Teacher amplification, not teacher replacement. Tools should make synchronous human time more effective. Tools that substitute for it should be evaluated with extreme caution.

•       Multilingual and equity-ready. Particularly in US and Canadian bilingual contexts (Spanish, French), tools that function only in English are not solutions. They are privileges.

•       Honest accountability. We ask every founder we work with the question no EdTech website currently answers: if students don't grow, what happens?

The transition ahead is not optional. The question is who will be in the room when the new infrastructure is designed. HierLearning's work is premised on the belief that district leaders—not platform companies, not venture capital, not AI labs—should be the primary stakeholders in that conversation.

We work directly with superintendents and school board leaders who are serious about turning their EdTech and AI investments into measurable learning gains. Not in the next product cycle. This school year.

You do not need to wait for the AI transition to finish to act. The structural forces are already moving. Here are the four questions that should be on every district leadership team's agenda in 2026. 

1. Which of our current EdTech tools depend on a habit loop to deliver value?

If a tool requires students to initiate engagement with a platform—to open an app, start a module, complete a sequence—it is structurally vulnerable. As AI agents become more capable, that initiation behaviour will decline. Audit your stack for this dependency.

2. What shared outcomes does our personalization infrastructure route toward?

Before AI personalization scales in your district, establish the common destinations. What does every student in this district need to know and be able to do? That framework cannot be delegated to a vendor.

3. Are we measuring the right things?

Time-on-platform, completion rates, and teacher satisfaction scores are internet-era metrics. They measure activity, not learning. If you cannot answer 'did students know more in June than they did in September,' you do not have an accountability framework. You have a reporting framework.

4. Which vendors in our stack can demonstrate outcomes—not just efficacy claims?

There is a difference between a randomized controlled trial published in 2018 and current pre/post data from a district like yours. Ask for the latter. If a vendor cannot produce it, that is your answer.

Social media did not intend to destroy collective sense-making. It was an emergent consequence of optimizing for engagement without asking what the engagement was for.

The EdTech industry is at the same inflection point. Full AI personalization without a shared outcomes framework will produce impressive individual engagement data and a quietly fragmented educational commons. The districts that recognize this risk now—and act on it—will not just survive the transition. They will lead it. 

HierLearning exists to make sure district leaders are in that room.

Your Personalization Strategy Is a Shared Curriculum Crisis Waiting to Happen:

You're at an inflection point that most EdTech vendors won't tell you about.

The tools your district runs today — the LMS, the adaptive platforms, the AI-powered dashboards — were built on internet-era logic: push content, track completion, optimize for engagement. That logic is collapsing. Not because the tools are bad. Because AI is dismantling the structural conditions that made them relevant.

The diagram maps what comes next and where the real risk lives.

The central node — the shared curriculum crisis — is the insight that matters most for district leaders. Full AI personalization, if adopted without a framework, doesn't just change how students learn. It erodes the common intellectual foundation that holds a classroom, a grade level, and a district together. Every student on their own algorithmically curated path is the educational equivalent of every person on their own social media feed. We've seen how that ends.

The fork in the diagram is your decision. One path is full personalization with no anchor — impressive short-term engagement data, fragmented long-term outcomes. The other is the thesis HierLearning is built on: personalized pathways toward shared, auditable destinations. Differentiated routes. Common outcomes. Measurable accountability.

The bottom of the diagram is practical. Four questions — habit loop audit, shared outcomes, right metrics, vendor accountability — are the ones to bring into your next procurement conversation. Any vendor who can't answer them cleanly is selling you the internet era with a new logo.

The closing line is the one worth sitting with: the question is not whether AI transforms K-12. It's whether that transformation gets designed by the people responsible for students, or by the platforms responsible to shareholders.

You'll hear it in every vendor pitch this year.

It sounds like progress. It might be the opposite.

Here's what "fully personalized learning" actually means when you pull back the curtain:

Every student on their own algorithmically curated path. Different texts, different examples, different conceptual sequences — optimized for individual engagement.

Sound familiar?

It should. Because that's exactly what social media promised. A feed tuned perfectly to you. Relevant. Personalized. Frictionless.

And we watched it quietly destroy the one thing that makes a shared society possible: common ground.

The tough question no vendor is asking:

Personalized toward what?

If a platform can't answer that — if there's no shared destination, no auditable common outcome, no moment where every student's individualized path converges — then you don't have a learning strategy.

You have a very expensive feed.

Here's what the research on learning actually tells us:

None of that happens in a fully siloed learning environment.

The real BS move is using personalization as a synonym for quality because:

👉 What do all of your students know and can do when June arrives?

Until a vendor can map their personalization engine to your district's shared outcomes — grade-level standards, literacy benchmarks, civic vocabulary — "AI-personalized" is just differentiation without accountability dressed up in a neural network.

Personalization is a route. A shared destination is the point.

Confusing the two is how EdTech keeps winning demos and losing classrooms.

Shout-out: A quiet shift is happening in the classrooms that are getting AI right — and it has nothing to do with the technology.

In districts where AI is actually moving outcomes, teachers aren't being replaced by a dashboard. They're being freed by one. The administrative drag — attendance workflows, sub coverage gaps, intervention tracking, reporting cycles — is getting absorbed by tools built for operations, not instruction. And what's left is what was always the point: a teacher in a room with students, with enough headspace to actually teach.

That's not a feature. That's a design philosophy.

The platforms worth watching aren't the ones promising to personalize every student's learning journey into a frictionless solo experience. They're the ones that recognize the classroom as the highest-leverage moment in a student's day — and build everything else around protecting it.

Translation: When AI handles the operational noise, teachers don't have to choose between managing the system and teaching the kid. The ones doing it right have figured out that the goal was never more technology in the classroom. It was more teacher in the classroom.

🏅 Recognition Sticker: "Cleared the Runway So Teachers Could Fly"

That’s all the unlock for today. Tune in next week

Stay awesome, you unlockers!

Want more? Order The Great Unlock here.