This episode explores the transformative yet complex integration of AI into education, revealing how new technologies are reshaping learning, teaching, and the very structure of educational institutions, offering key parallels for innovation in other complex sectors.
1. Introduction to AI in Education and Expert Insights
- Justine and Olivia introduce the week's topic: the burgeoning trend of AI in education, particularly its visibility on social media.
- They bring in Zach, an A16Z colleague with extensive experience in edtech, including operating, investing, and founding an education technology company focused on high school computer science.
- Zach's background includes roles at General Atlantic, with investments in Quizlet and Duolingo, and building a company later sold to an education-focused rollup, providing him a deep understanding of both consumer and institutional edtech.
2. The Evolving School Response to AI
- Olivia notes the initial significant usage of AI like ChatGPT by students, which was surprisingly larger than non-student use in its first year. This led to an immediate and severe backlash, including the rise of AI detection tools.
- Zach addresses the evolution of schools' stance on AI, stating, "I think we're super far away from that moment [of banning AI] which I think is really, really strong now."
- K-12 Education: While some skepticism remains, approximately 80% of districts now have generative AI teams and earmarked budgets for procuring new AI technologies. Generative AI refers to artificial intelligence capable of creating new content, such as text, images, or in this case, educational materials. There's ongoing discussion about the necessity of a "teacher in the loop."
- Higher Education: This sector is leading AI adoption. Companies like Claude and OpenAI are launching education-specific platforms and piloting with universities. Higher ed institutions recognize AI as an essential tool for future careers.
- Zach highlights that the rapid progress in AI adoption within 18 months is impressive for the education sector, which is typically slow to adopt new technologies (e.g., the slow move to cloud). Some institutions, like Ohio State, are even mandating AI usage in their curriculum.
- Strategic Implication: The rapid yet cautious adoption curve in education, a traditionally slow-moving sector, offers insights for Crypto AI ventures targeting established industries. Understanding the initial resistance, the role of dedicated internal teams, and the gradual shift from periphery to core integration is crucial.
3. AI Adoption: Teachers Leading the Way
- Justine outlines the diverse segments of the education market: public, private, charter, homeschooling, supplementary products, higher ed, and adult/corporate learning.
- Zach reveals a surprising trend: teachers, not students or adult learners, are the primary adopters of AI tools, often paying out-of-pocket.
- He notes, "It's teachers who are willing to kind of pay and use this in their every single day workflow."
- AI significantly reduces the administrative burden teachers face (grading, feedback, curriculum development), which constitutes about 90% of the work they dislike.
- Tools like Magic School, reportedly used by 50% of US teachers with over 5 million users, demonstrate this bottoms-up adoption.
- While students use AI for homework, and adult learning platforms see some AI feature benefits, a native AI player hasn't dominantly emerged in adult education, possibly due to distribution challenges or the effectiveness of layering AI onto existing strong pedagogical frameworks.
- Actionable Insight: The "teacher-led" adoption model in edtech, driven by clear ROI in productivity, suggests that Crypto AI solutions targeting professional workflows must offer immediate, tangible benefits to individual users, even before institutional buy-in.
4. Defining and Measuring "What's Working" in AI Edtech
- Olivia raises a critical question: what truly signifies success in AI edtech—high usage (like ChatGPT for homework) or actual improvements in learning, retention, and memory?
- Zach distinguishes between investor perspectives and learning outcome perspectives on "what's working."
- Investor Metrics: Focus on retention and engagement, particularly cohorted engagement (e.g., how many days a week a user cohort uses the app over time). This can indicate if the product is becoming integral to learning, not just a tool for quick homework completion. Monthly retention is often more telling than weekly or yearly due to academic cycles.
- Learning Outcomes: Measuring direct impact on learning is challenging. AI is still largely at the periphery of core education, making it difficult to isolate its impact on standardized test scores, which require multi-year data.
- Zach states, "AI is still at the periphery of education...we don't actually know what the outsized impact is."
- While some research papers show test improvements with AI-instructed courses, these are not yet large-scale, nationwide studies.
- Strategic Consideration: For Crypto AI researchers and investors, this highlights the universal challenge of defining and measuring true "value" beyond surface-level engagement. Developing robust, context-specific benchmarks for efficacy is key, especially in fields where outcomes are complex and long-term.
5. Alpha School: A Glimpse into AI-Centric Education
- Justine introduces Alpha School, run by Mackenzie Price, which has gained viral social media attention for its AI-driven educational model. Students reportedly spend two hours a day on AI tutor-led classwork and the rest on self-directed projects.
- Zach views Alpha School as an "innovation lab" for education.
- Its status as a well-funded private school ($40,000 tuition) with a self-selecting, tech-receptive parent/student base allows it to experiment aggressively with AI.
- "I think what Alpha School is showing us is like if you kind of turn full tilt on AI in education...you can kind of figure out what's working," Zach explains.
- Alpha School students have reportedly achieved top 1-2% national rankings, demonstrating significant outcomes.
- However, Zach cautions that this model's success is in a privileged environment, and commercialization challenges (integration, cost, software literacy in typical schools) remain for broader adoption.
- Implication for Crypto AI: Alpha School's model, while resource-intensive, showcases the potential of fully integrating advanced technology. Crypto AI projects might consider "lighthouse" initiatives or sandboxed environments to demonstrate transformative potential, even if widespread adoption faces hurdles.
6. Will AI Replace Teachers?
- Olivia poses the controversial question: will AI software replace human teachers, especially given the high cost of personnel versus software?
- Zach believes teacher replacement is unlikely or very far off. There's an existing teacher shortage, and current AI applications primarily augment teacher productivity (e.g., generating worksheets) rather than directly interacting with students in a learning environment.
- "They're allowing a teacher to be a lot more productive...but it doesn't mean that the students are interacting with AI in a learning environment," Zach clarifies.
- He distinguishes between AI helping teachers create existing educational assets (worksheets) versus AI delivering entire learning units or immersive experiences (e.g., conversing with a historical avatar). The latter is still far from mainstream.
- The focus is on reducing teacher burnout and improving efficiency, not replacing the human element of teaching. Active teaching might decrease with AI assistance, but not disappear.
- Strategic Insight: The "AI as augmentation, not replacement" theme is prevalent across many industries. Crypto AI solutions should focus on empowering existing roles and workflows, rather than aiming for full automation where human interaction and oversight remain critical.
7. The Rise of AI-Generated Edutainment and New Learning Modalities
- Olivia and Justine discuss the explosion of free, AI-generated educational content on platforms like TikTok and Instagram, which looks very different from traditional edtech like Khan Academy.
- Examples include NotebookLM (a tool for interacting with your notes and sources, here used to turn a textbook into a podcast) and deepfake celebrity explanations of complex topics (e.g., Sydney Sweeney explaining physics on the "Unlock Learning" Instagram account). Deepfakes are synthetic media in which a person in an existing image or video is replaced with someone else's likeness using AI.
- Zach notes these videos are engaging and effectively teach detailed, technical topics, improving in quality and complexity (e.g., Drake and Sydney Sweeney discussing 3D vectors).
- He observes, "They feel like brain rot content, but they're actually the complete opposite."
- This trend signifies a shift from boxing learners into one "type" (visual, audio, etc.) to allowing them to choose modalities based on the topic and their learning goals.
- Emerging Trend: The diversification of content delivery through AI points to a future where learning is highly personalized by format and style. For Crypto AI, this suggests opportunities in creating novel interfaces and experiences for complex information, moving beyond traditional documentation or UIs.
8. The Challenge of School Adoption for Innovative AI Content
- Zach points out a disconnect: while engaging AI content like historical figure chatbots exists, schools often adopt more rudimentary AI tools (worksheet generators) first due to ease of integration into existing practices.
- "In order to get school adoption they have to wedge in with these kind of like more rudimentary tools...And then once you do that it's not so easy to get people to use the other mode because it's a new form of education," he explains.
- Usage of innovative AI learning experiences remains low in schools compared to tools that simply make existing tasks easier. A significant professional development (PD) movement is needed to help teachers integrate AI into the classroom, not just their workflow.
- Actionable Implication: Crypto AI innovators should anticipate that even groundbreaking solutions might need a "Trojan horse" strategy—entering a market with tools that address immediate, familiar pain points before introducing more transformative, but less familiar, applications.
9. The Role of Textbook Companies and Parental Demands
- Justine highlights how AI-generated content can separate content creation from content delivery, optimizing both.
- Zach suggests textbook companies are key gatekeepers, controlling much of the content in classrooms. Their willingness to partner with AI companies or innovate internally will significantly shape AI's role.
- "It's a weird kind of power dynamic right now," Zach notes, regarding whether textbook companies will view AI as a threat or an extension.
- Olivia raises the question of whether parents, seeking better outcomes, will increasingly opt for AI-directed education.
- Zach believes parents prioritize outcomes. He cites an early-stage AI reading company promising to get 3-4 year olds to a third-grade reading level in months for $500/month, which attracted initial sign-ups.
- The appeal of AI education is socioeconomically dependent. For some, it's a powerful alternative to passive screen time; for others with access to expensive private tutors, the value proposition of AI is different.
- The controllability of LLMs (Large Language Models)—AI systems trained on vast amounts of text data to understand and generate human-like language—is attractive to parents who want to customize their child's learning.
- Strategic Consideration: The interplay between incumbents (textbook companies) and new AI-driven solutions, alongside evolving consumer (parental) demand for measurable outcomes and customization, mirrors dynamics in many tech disruptions. Crypto AI ventures must navigate these forces.
10. Future Outlook: AI in Edtech in the Next Year
- Olivia asks Zach for his predictions for AI in edtech over the next 12 months.
- Zach anticipates:
- Significant progress in higher education's use of AI.
- Clarity on whether large model companies will suffice or if smaller, specialized application companies built on top will thrive.
- A move from AI as a peripheral tool to more integrated classroom use, potentially unlocked by advances in real-time voice AI.
- He states, "I think the next 12 months will show us a lot. I don't know if it will necessarily like birth a massive company in education...but I think education will look fairly similar though."
- Justine expresses a desire to see the first fully AI teacher influencer (not a deepfake celebrity) and more apps enabling personalized learning speeds, moving away from the one-pace-fits-all classroom model.
- Zach agrees that LLMs are capable of teaching but require better engagement infrastructure (avatars, voice). "We're at the moment where I think LLMs are trusted as a learning partner. Now we have to build experiences around them."
- Future Watchpoint: The evolution from foundational AI models to engaging, specialized applications is a key trend. Crypto AI investors should monitor how this "experience layer" develops, as it's crucial for user adoption and unlocking the true potential of underlying technologies.
Conclusion: AI's Educational Evolution Offers Broader Tech Parallels
The discussion reveals AI's gradual but profound impact on education, driven by teacher needs and new content forms. For Crypto AI investors and researchers, this highlights the importance of user-centric design, demonstrable ROI, and navigating incumbent structures when introducing transformative technologies into established, complex systems.
