Irina Jurenka, Research Lead at Google DeepMind, and Muktha Ananda, Engineering Leader in Learning and Education at Google, share their insights on developing LearnLM, a large language model tailored for education. They delve into the intricacies of fine-tuning AI to enhance pedagogical effectiveness, explaining how they measure learner outcomes and the challenges of creating an engaging AI tutor. The conversation highlights the delicate balance between emotional engagement and learning efficiency, showcasing a multidisciplinary approach to innovation in educational technology.

Ben Kornell, Managing Partner at Common Sense Growth Fund and co-founder of Edtech Insiders, dives into the intricacies of AI tutoring. He differentiates between AI-powered search and genuine tutoring, emphasizing how AI can enhance human interaction. The conversation explores age-specific needs, with younger students benefiting from personal connections while older ones seek independence. Ethical concerns, such as bias and dependency, are also discussed, alongside exciting future developments that may make AI tutoring feel almost like science fiction!

Rob Wong, the Product Lead for LearnX at Google, dives into the innovative intersection of AI and education. He sheds light on the challenges of modeling learner profiles and the importance of balancing user needs with educational value. The discussion showcases exciting AI tools like YouTutor and Learning Coach Gem, emphasizing personalization in learning experiences. Wong also explores the shift to an 'AI first' approach and the ongoing efforts to fine-tune AI for supportive, engaging education. Prepare for insights into the future of learning!

Ryan Baker, a Professor at the University of Pennsylvania and Director of the Penn Center for Learning Analytics, dives into the fascinating world of big data and algorithmic bias in education. He highlights how educational data mining can enhance learner engagement and outcomes. The discussion reveals the nuances of algorithmic bias, its societal implications, and why tailored approaches are necessary to ensure fairness. Moreover, Baker debunks myths about AI in education, advocating for a balanced integration that supports educators.

In this short episode, Libby and Owen discuss OpenAI’s new model for advanced reasoning, o1. They talk about its new capabilities and strengths, and what they think about its significance for education after an initial play around. They talk through the benefits of ‘think aloud’ versus ‘think before you speak’ approaches in education, and how this relates to o1. Links:OpenAI’s announcement about o1Join us on social media: BOLD (@BOLD_insights), Libby Hills (@Libbylhhills) and Owen Henkel (@owen_henkel) Listen to all episodes of Ed-Technical here: https://bold.expert/ed-technical Subscribe to BOLD’s newsletter: https://bold.expert/newsletter Stay up to date with all the latest research on child development and learning: https://bold.expert Credits: Sarah Myles for production support; Josie Hills for graphic design

Craig Barton, Head of Education at Eedi and a math podcast host, joins Simon Woodhead, Director of Research at Eedi, to dive deep into educational misconceptions. They discuss how AI can enhance understanding and address errors in math education. Listeners will learn about the importance of identifying misconceptions using diagnostic questions and how AI integration can support teachers in overcoming real-world challenges. The conversation emphasizes both the potential and limitations of AI in enriching student learning experiences.

In this short, Libby and Owen discuss recent research from Anthropic looking at sycophancy – the tendency to agree with users – in large language models (LLMs), and key research from educational psychology about how important feedback is for learning. Libby and Owen connect the two papers and explore why sycophancy is especially a problem when it comes to using LLMs for educational purposes. Links:Anthropic paper on sycophancy in language models John Hattie and Helen Timberley’s paper, The Power of Feedback Join us on social media: BOLD (@BOLD_insights), Libby Hills (@Libbylhhills) and Owen Henkel (@owen_henkel) Listen to all episodes of Ed-Technical here: https://bold.expert/ed-technical Subscribe to BOLD’s newsletter: https://bold.expert/newsletter Stay up to date with all the latest research on child development and learning: https://bold.expert Credits: Sarah Myles for production support; Josie Hills for graphic design

The discussion starts with a humorous detour about a political figure's teaching legacy. Libby and Owen dive into the world of automated lesson planning tools like Khanmigo and Magic School, exploring their features and benefits. They explain the significance of structured lesson plans and how these tools differ from traditional resources. The challenges of balancing efficiency with personalized teaching emerge, along with the importance of reflective practices. Finally, there's a light-hearted look at memes related to their favorite figure, juxtaposing education with entertainment.

In this short, Libby and Owen discuss a recent paper that has generated interest and discussion called ‘Generative AI Can Harm Learning’. The paper presents the findings from a thought-provoking study of nearly 1,000 students in Turkey. The study tested the effects of giving students access to two different versions of GPT-4 while studying math: one was essentially ChatGPT and the other was a version of GPT-4 that had been tailored for tutoring with a thin prompt wrapper – so it didn’t just give students the answer. The main finding (that the title is based on) is that access to generic ChatGPT had a negative effect on students’ math test results, versus the control group who studied with no access to a chatbot. Not everyone agrees that the results justify the somewhat dramatic title, or that the title reflects the most interesting findings from the study. Listen in to see what Libby and Owen think. The ‘Generative AI Can Harm Learning’ paper can be found here.Join us on social media: BOLD (@BOLD_insights), Libby Hills (@Libbylhhills) and Owen Henkel (@owen_henkel) Listen to all episodes of Ed-Technical here: https://bold.expert/ed-technical Subscribe to BOLD’s newsletter: https://bold.expert/newsletter Stay up to date with all the latest research on child development and learning: https://bold.expert Credits: Sarah Myles for production support; Josie Hills for graphic design

This episode, Owen and Libby speak to Carmen Strigel, Senior Director of Education Technology at RTI, a non-profit global research organisation. Carmen has been the driving force behind a number of successful EdTech products built and used in low resource settings. Carmen tells Owen and Libby about Tangerine, data collection software used in more than 60 countries, and Loquat, a machine learning tool that provides feedback to teachers on their classroom talk. This episode builds on earlier interviews this season about voice AI with Alyssa van Kamp from TeachFX and Peter Foltz from the Institute of Cognitive Science at the University of Colorado Boulder. Carmen tells Libby and Owen aboutvoice AI in low resource settings. They also explore how contextual factors, like different pedagogies and classroom settings, influence EdTech product design in low resource settings. Carmen and Owen connect over their shared passion for assessment - watch this space for the ‘Formative Assessment Fanboy’ t-shirts proposed in the episode. Links:Carmen Strigel’s biography and publicationsInformation about TangerineInformation about LoquatJoin us on social media: BOLD (@BOLD_insights), Libby Hills (@Libbylhhills) and Owen Henkel (@owen_henkel) Listen to all episodes of Ed-Technical here: https://bold.expert/ed-technical Subscribe to BOLD’s newsletter: https://bold.expert/newsletter Stay up to date with all the latest research on child development and learning: https://bold.expert Credits: Sarah Myles for production support; Josie Hills for graphic design