Google is testing LearnLM in Google Classroom to simplify lesson planning and help teachers find tailored content, while also using LearnLM to power Circle to Search on Android for solving math and physics problems, and a YouTube tool that enables users to interact with academic videos by asking questions, getting explanations, and taking quizzes. (Source: Image by RR)

Google Partners with Top Educational Institutions to Expand Impact of LearnLM

Google has introduced LearnLM, a new family of generative AI models designed for educational purposes, developed in collaboration between DeepMind AI and Google Research. As noted in, built on Google’s Gemini models, LearnLM aims to tutor students conversationally across various subjects and is already integrated into multiple Google products like YouTube, Google Search, and Google Classroom. During a keynote at Google’s I/O 2024 developer conference, James Manyika, SVP of Research, Technology and Society, highlighted that LearnLM is grounded in educational research to create more personalized and engaging learning experiences.

The pilot program in Google Classroom is exploring how LearnLM can simplify and enhance lesson planning for educators, helping them discover new ideas and tailored content for specific student needs. Additionally, LearnLM powers features like Circle to Search on Android, aiding in solving math and physics problems, and a YouTube tool that allows users to ask questions and take quizzes based on academic videos. In the coming months, LearnLM will also enable users to create custom chatbots in Google’s Gemini apps, providing study guidance and practice activities.

Google plans to extend LearnLM’s capabilities beyond its products through partnerships with institutions like Columbia Teachers College, Arizona State University, NYU Tisch and Khan Academy. Manyika emphasized that this initiative marks a new chapter for learning and education at Google, leveraging generative AI to make information and knowledge universally accessible and useful. However, a technical paper on LearnLM’s development notes some issues, such as an unencouraging tone and difficulty recognizing correct answers, similar to other generative AI models.

The paper also highlights concerns about LearnLM’s potential for generating inaccurate information, urging further evaluation before broad application. Despite these challenges, Google assures that it has conducted thorough evaluations for its apps to ensure safety and effectiveness for students and educators, aiming to address these issues moving forward.