Emma Dorn | McKinsey & Company

Emma Dorn | McKinsey & Company

Global Tech, Local Impact: Unlocking Classroom Bottlenecks with Emma Dorn How can we let data drive student success? On this episode, host Jeremy Singer talks with Emma Dorn, a Senior Knowledge Expert at McKinsey & Company [https://www.mckinsey.com/], about how school systems improve and sustain progress at scale. While the US debate focuses on whether to use technology, the Global South innovates out of necessity due to resource constraints. From solar tablets in Malawi to WhatsApp coaching in Pakistan, Emma shares how to solve the core instructional bottleneck. Emma outlines five case studies from around the world where technology solves real instructional problems:  • First, design for the instructional bottleneck, not the tech [http://globalpartnership.org/blog/malawi-strengthening-edtech-evidence-community-perspectives]. In Malawi, where class sizes reach 150, Imagine Worldwide deploys offline, solar-powered tablets running adaptive software, doubling literacy and math fluency.  • Second, technology designed for teachers can be more transformative than direct-to-student devices [https://www.mckinsey.com/industries/education/our-insights/drivers-of-student-performance-insights-from-europe]. McKinsey's analysis of global PISA data shows tech in the hands of teachers adds a year or two of learning, while student devices reduce outcomes due to distraction. New Globe utilizes teacher tablets to deliver structured lesson plans offline.  • Third, technology works better when aligned with the curriculum [https://mentulabs.com/en/]. Mentu's AI assistant in Colombia and the Dominican Republic maps to local curricula to scaffold lower-skill teachers.  • Fourth, technology is most powerful when scaling proven interventions like teacher coaching [https://taleemabad.com/]. In Pakistan, Talimabad created a WhatsApp-native digital coach analyzing classroom audio to give immediate feedback.  • Fifth, think beyond the Scantron [https://www.pratham.org/about/teaching-at-the-right-level/]. Using the Teaching at the Right Level (TaRL) methodology, Pratham and Wadhwani AI developed oral reading fluency assessments operating on $50 cell phones, reducing assessment time by 50% across millions of learners.Listen in to learn how a global perspective can provide unique insights into how to leverage emerging technologies to solve real world educational challenges at scale. Research and Links: Malawi Scale-Up: globalpartnership.org/blog/malawi-strengthening-edtech-evidence-community-perspectives [http://globalpartnership.org/blog/malawi-strengthening-edtech-evidence-community-perspectives] McKinsey PISA Study: https://www.google.com/search?q=https://mckinsey.com/industries/education/our-insights/drivers-of-student-performance-insights-from-europehttps://www.mckinsey.com/industries/education/our-insights/drivers-of-student-performance-insights-from-europe [https://www.mckinsey.com/industries/education/our-insights/drivers-of-student-performance-insights-from-europe] Structured Pedagogy: https://www.wwhge.org/resources/implementing-structured-pedagogy-programmes-effectively-at-scale/ [https://www.wwhge.org/resources/implementing-structured-pedagogy-programmes-effectively-at-scale/] New Globe Impact: https://newglobe.education/impact/ [https://newglobe.education/impact/] Taleemabad Profile: https://taleemabad.com/ [https://taleemabad.com/] Mentu AI: https://mentulabs.com/en/ [https://mentulabs.com/en/] Pratham TaRL: https://www.pratham.org/about/teaching-at-the-right-level/ [https://www.pratham.org/about/teaching-at-the-right-level/] Wadhwani AI: https://www.wadhwaniai.org/ [https://www.wadhwaniai.org/] Join the Conversation: Subscribe, like, and follow The Education Equation wherever you get your podcasts. Time Stamps: 00:00 Show Introduction 00:38 Meet Emma Dorn 02:03 Why Look Globally 04:14 Constraints Drive Innovation 06:20 Lesson One Bottlenecks 12:21 Lesson Two Teacher Tech 19:00 Lesson Three Curriculum Alignment 23:37 Lesson Four Scaling What Works 29:28 Lesson Five Oral Assessments 34:18 Big Picture Takeaways 35:42 Rapid Fire Questions 37:40 Five Year Vision 38:51 Closing Thanks

Ethan Mollick | The Wharton School

AI and the Future of Learning with Ethan Mollick Jeremy Singer sits down with Ethan Mollick, professor at the University of Pennsylvania's Wharton School and author of Co-Intelligence: Living and Working with AI [https://www.penguinrandomhouse.com/books/741805/co-intelligence-by-ethan-mollick/], a New York Times bestseller and a best book of the year from The Economist and Financial Times. Mollick is a defining voice in the AI space, testing generative tools in real-world classrooms and authoring the widely read newsletter One Useful Thing [https://www.oneusefulthing.org/]. He is the Ralph J. Roberts Distinguished Faculty Scholar, Rowan Fellow, and Associate Professor at the Wharton School, where he directs the Generative AI Lab and studies the effects of AI on work, entrepreneurship, and education. He was named one of TIME Magazine's Most Influential People in Artificial Intelligence.  Despite the rapid pace of development, Mollick argues that the AI we use today is the worst it will ever be, necessitating a proactive shift from denial to engagement. In this episode, he explains why the "homework apocalypse" is already here and how the traditional writing assignment has been fundamentally disrupted. The discussion moves beyond simple prompt engineering to explore the "jagged frontier" of AI capability, where the technology excels at complex tasks while occasionally stumbling on simple ones. Mollick and Singer break down the varying impacts of AI on human performance, describing it as a leveler for those with skill gaps and a "king-maker" for experts who can leverage the tool to achieve massive productivity gains. The conversation also tackles the existential questions of human agency and the future of assessment. Mollick advocates for a "flipped classroom" model where AI serves as an infinitely patient universal tutor outside of school, leaving class time for active learning, debates, and creative projects. He and Singer discuss testing in the age of AI and how educators can teach students to develop "durable skills" like taste, judgment, and a unique personal style . By viewing AI as a teammate rather than just a tool, Mollick suggests that teachers can reclaim hours of their week and focus on the deeply human aspects of education. Featured References: * Book: Co-Intelligence: Living and Working with AI [https://www.penguinrandomhouse.com/books/741805/co-intelligence-by-ethan-mollick/] by Ethan Mollick * Newsletter: One Useful Thing [https://www.oneusefulthing.org/] by Ethan Mollick * Book: The Diamond Age [https://www.amazon.com/Diamond-Age-Illustrated-Primer-Spectra/dp/0553380966] by Neal Stephenson (The "Young Lady’s Illustrated Primer") * Research: Wharton AI Labs [https://gail.wharton.upenn.edu/] on student performance and AI interaction * Data: Walton Family Foundation [https://www.waltonfamilyfoundation.org/ai-in-the-classroom] survey on teacher AI usage and time-savings * Case Studies: AI-driven educational outcomes and tutor studies in Taiwan [https://www.nature.com/articles/s41598-025-97652-6] and outsourcing academic writing to Kenya [https://www.theguardian.com/education/2025/sep/14/kenya-shadow-scholars-paid-to-write-essays-for-uk-students] Episode Timestamps: 00:00 — Introduction to Ethan Mollick and the pace of AI evolution 02:26 — The "homework apocalypse" and the shift in educational thinking 04:45 — Why today’s AI is the worst version we will ever use 09:15 — Lessons from the 1970s calculator debate for modern curricula 13:55 — AI literacy vs. a fundamental reshaping of human interaction 17:30 — Developing taste and agency as essential "durable skills" 19:00 — The Leveler, Elevator, and King-maker effects on human skill 29:10 — The promise of the universal tutor to close equity gaps 32:55 — Using AI to enable the flipped classroom and active learning 36:55 — Shifting from "tool" to "teammate" in teacher workflows 41:50 — Why AI detectors fail and the move toward in-class assessment 45:45 — Rapid Fire: Retirement of AI literacy, Neal Stephenson, and the importance of humanities 48:45 — A three-year vision for positive change in education Follow, Like, and Share The Education Equation [https://theeducationequation.org/] wherever you get your podcasts.

Daniel Willingham | Cognitive Psychology at the University of Virginia

Cognitive Science in the Classroom with Dan Willingham Jeremy Singer sits down with Dr. Daniel Willingham, professor of psychology at the University of Virginia and author of Why Don’t Students Like School?: A Cognitive Scientist Answers Questions About How the Mind Works and What It Means for the Classroom [https://www.amazon.com/Why-Dont-Students-Like-School/dp/1119715660]. Despite decades of robust research into how the human brain learns, a significant gap remains between laboratory findings and daily classroom instruction. In this episode, Dr. Willingham shares the origin story of his career pivot from memory researcher to educational translator following a transformative 2001 conference with teachers in Nashville. The discussion moves beyond the hype of educational trends to explore the fundamental constraints of the human mind. Willingham and Singer break down why students are naturally averse to thinking, explaining that the brain is designed to save effort by relying on memory rather than slow, hazardous problem-solving. They discuss why spacing—revisiting content over time—is scientifically superior for retention but often resisted by students due to its impact on immediate motivation. The conversation also tackles the modern challenges of AI and declining reading scores. Willingham argues that the issue isn't necessarily a lack of leisure reading among youth, but rather a decrease in the rigor of texts and expectations within schools. He advocates for knowledge-rich curricula that sensibly sequence factual, procedural, and conceptual information to provide the necessary engine for critical thinking. Featured References: Book: Why Don't Students Like School? [https://www.amazon.com/Why-Dont-Students-Like-School/dp/1119715660] by Daniel Willingham Book: The Reading Mind [https://www.amazon.com/Reading-Mind-Cognitive-Approach-Understanding/dp/1119301378] by Daniel Willingham Book: Cultural Literacy [https://www.amazon.com/Cultural-Literacy-Every-American-Needs/dp/0394758439] by E.D. Hirsch Research: Sam Wineburg (Stanford University) on digital literacy and vetting information [https://ed.stanford.edu/news/digital-literacy-ai-era-part-1] Data: American Time Use Survey (Bureau of Labor Statistics) [https://www.bls.gov/tus/] Episode Timestamps: 00:00 — Introduction to Dr. Daniel Willingham and the gap between science and classrooms 03:00 — The Nashville story: Why Willingham pivoted to education 09:00 — The science of spacing: Why laboratory findings don't always translate to classrooms 13:50 — Advice for superintendents: Choosing CogSci-aligned curricula 22:45 — The core thesis: Why the brain is set up to save you from thinking 33:45 — Defining factual, procedural, and conceptual knowledge 41:50 — AI in education: Generalizable skills vs. domain-specific knowledge 51:40  The Reading Mind: Addressing declining scores and the importance of rigor 01:03:50 — Rapid Fire: Retirement of learning styles, the importance of probability, and the future of teacher education Follow, Like, and Share The Education Equation [https://educationequation.collegeboard.org/] wherever you get your podcasts.

Jennie Magiera | Education Impact at Google

Jeremy Singer, president of the ⁠College Board⁠ [https://collegeboard.org], is joined by Jennie Magiera, Google’s Global Head of Education Impact, to discuss the transformative role of AI in schools. Magiera, who began her career as a math teacher and tech skeptic at Chicago Public Schools, explains her shift toward viewing technology as a powerful vehicle for educational equity when it is used to addressspecific instructional challenges. The discussion highlights ⁠LearnLM⁠ [https://cloud.google.com/solutions/learnlm], Google’s family of AI models fine-tuned specifically for education and grounded in five core learning science principles: active learning, managing cognitive load, adapting to learners, stimulating curiosity, and deepening metacognition. Magiera shares ⁠a study [https://storage.googleapis.com/deepmind-media/LearnLM/learnLM_nov25.pdf]⁠ showing that students who used LearnLM for short tutoring sessions experienced a 5.5% performance boost on novel topics compared to those working with human tutors alone. ⁠A six-month Northern Ireland pilot⁠ [https://blog.google/innovation-and-ai/models-and-research/google-deepmind/ai-classroom-northern-ireland/] further demonstrates AI's impact, with teachers reporting an average of 10 hours saved per week on administrative tasks like lesson planning, risk assessments, and parent outreach. This efficiency provides critical breathing room for educators, helping to reduce stress and prevent burnout while allowing them to reinvest time into professional learning and deeper student engagement. The episode concludes with an exploration of durable skills like empathy and critical thinking, which Magiera argues will only increase in importance as AI handles more routine technical tasks. Resources Mentioned in This Episode: ⁠Google for Education Communities and Tools⁠ [https://www.googleforeducommunity.com/] ⁠Switch by Dan and Chip Heath⁠ [https://www.amazon.com/Switch-Change-Things-When-Hard/dp/0385528752] ⁠Weekend Language by Andy Craig and Dave Faustino⁠ [https://a.co/d/0gbrFEGs] ⁠OECD AI Capability Indicators Report⁠ [https://www.oecd.org/en/publications/introducing-the-oecd-ai-capability-indicators_be745f04-en/full-report.html] Google for Education Research in Northern Ireland [https://blog.google/innovation-and-ai/models-and-research/google-deepmind/ai-classroom-northern-ireland/] Google Research with Eedi Labs about AI Tutoring [https://eedi-labs-67qwnm2lxu4blqi5.webflow.io/news/new-exploratory-research-from-eedi-and-google-deepmind-reveals-human-in-the-loop-ai-tutoring-outperforms-human-only-support] Episode Timestamps: 00:00 — Jeremy Singer introduces the Education Equation 01:05 — Meet Jennie Magiera, Global Head of Education Impact at Google 05:43 — From tech skeptic to innovation leader 08:14 — The Lead with the Need philosophy 18:10 — Understanding LearnLM and learning science principles 22:30 — Research on AI tutoring and student performance boosts 29:25 — Saving 10 hours a week: The Northern Ireland Pilot 37:45 — Durable skills and what makes us uniquely human 43:15 — Rapid Fire: Book recommendations and the importance of empathy 48:10 — Hopes for education in the next three years Like, Share, and Follow ⁠The Education Equation⁠ [https://educationequation.org] wherever you get your podcasts.

Sal Khan | Khan Academy

In this episode of The Education Equation [https://educationequation.collegeboard.org/], host Jeremy Singer, President of the College Board [https://collegeboard.org] talks with Sal Khan, founder and CEO of Khan Academy [https://khanacademy.org], about the evolution of their decade-long partnership and the transformative role of AI in the classroom. They reflect on the early days of providing free, world-class SAT practice to tens of millions of students and share the "beast mode" origins of Sal’s instructional videos. The conversation moves into the launch of a new collaboration between the College Board and Schoolhouse.world [https://schoolhouse.world], which provides free, peer-to-peer SAT tutoring to help students stay accountable and motivated. Sal discusses insights from his book, Brave New Words: How AI Will Revolutionize Education (And Why That’s A Good Thing) [https://www.penguinrandomhouse.com/books/740806/brave-new-words-by-salman-khan/], and describes his first experience with GPT-4 as a "benevolent alien moment" that signaled a new era for educational technology. He explains how AI tools like Khanmigo serve as Socratic tutors that nudge students toward answers without doing the work for them, while also acting as powerful assistants for teachers by handling administrative tasks and lesson planning. Despite the rapid growth of AI, Sal argues that teaching remains one of the most human-centric and secure professions in the modern world. The episode also covers the ethics of AI in schools, where Sal advises educators to avoid fool's errands like AI detection and instead embrace transparency and in-class assessments. He shares personal anecdotes about AI’s potential to support mental health and family dynamics, illustrating how these tools can provide meaningful advice for young people. Finally, Jeremy and Sal envision a future where technology makes classrooms more joyous and interactive, allowing teachers to focus on student engagement rather than paperwork. Sal's book recommendation: The Foundation Series [https://a.co/d/3iq7Vk7] by Isaac Asimov Like, Share, and Follow The Education Equation wherever you get your podcasts. Timestamps (00:00) – Introduction: Jeremy Singer and the mission of The Education Equation (02:18) – The origin story of the Khan Academy and College Board partnership (05:47) – Sal’s "beast mode" process for creating instructional content (08:35) – Data and evidence: The impact of free personalized practice (12:12) – Evolution from supplemental tool to district-wide integration (15:58) – The birth of Schoolhouse.world and the future of peer tutoring (22:20) – Balancing pedagogy with engagement: The Duolingo vs. Khan Academy approach (24:49) – The "Benevolent Alien" moment: First contact with GPT-4 (29:50) – Mastery learning and addressing "unfinished learning" with AI (33:00) – Guardrails: Ensuring AI helps students learn without cheating (37:25) – Vision for the AI-enabled classroom of the future (40:55) – AI's role in mental health and family dynamics (45:43) – Advice for educators on using AI ethically and responsibly (47:50) – Rapid Fire: Educational buzzwords, favorite books, and the "AP Law" idea (51:00) – Closing thoughts and where to follow