Google DeepMind: The Podcast

Cooperation is at the heart of our society. Inventing the railway, giving birth to the Renaissance, and creating the Covid-19 vaccine all required people to combine efforts. But cooperation is so much more. It governs our education systems, healthcare, and food production. In this episode, Hannah meets the researchers working on cooperative AI, and hears about their work and influences from the famous American psychologist - and pigeon trainer -  BF Skinner to the strategic board game Diplomacy. For questions or feedback on the series, message us on Twitter @DeepMind or email podcast@deepmind.com Interviewees: DeepMind’s Thore Graepel, Kevin Mckee, Doina Precup & Laura Weidinger CreditsPresenter: Hannah FrySeries Producer: Dan HardoonProduction support: Jill AchinekuSounds design: Emma BarnabyMusic composition: Eleni ShawSound Engineer: Nigel AppletonEditor: David PrestCommissioned by DeepMind Thank you to everyone who made this season possible!  Further reading: Machines must learn to find common ground, Nature: https://www.nature.com/articles/d41586-021-01170-0Introduction to Reinforcement Learning, DeepMind: https://www.youtube.com/watch?v=2pWv7GOvuf0B.F. Skinner, Wikipedia: https://en.wikipedia.org/wiki/B._F._SkinnerThe Tragedy of the Commons, Wikipedia: https://en.wikipedia.org/wiki/Tragedy_of_the_commonsStaving Off The Ultimate Tragedy Of The Commons, Forbes: https://www.forbes.com/sites/georgebradt/2021/11/02/staving-off-the-ultimate-tragedy-of-the-commons-by-making-better-complex-decisions-cooperatively-in-glasgow/Understanding Agent Cooperation, DeepMind: https://deepmind.com/blog/article/understanding-agent-cooperationThe emergence of complex cooperative agents, DeepMind: https://deepmind.com/blog/article/capture-the-flag-science Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

Hannah explores the potential of language models, the questions they raise, and if teaching a computer about language is enough to create artificial general intelligence (AGI). Beyond helping us communicate ideas, language plays a crucial role in memory, cooperation, and thinking – which is why AI researchers have long aimed to communicate with computers using natural language. Recently, there has been extraordinary progress using large-language models (LLM), which learn how to speak by processing huge amounts of data from the internet. The results can be very convincing, but pose significant ethical challenges.  For questions or feedback on the series, message us on Twitter @DeepMind or email podcast@deepmind.com.  Interviewees: DeepMind’s Geoffrey Irving, Chris Dyer, Angeliki Lazaridou, Lisa-Anne Hendriks & Laura Weidinger  CreditsPresenter: Hannah FrySeries Producer: Dan HardoonProduction support: Jill AchinekuSounds design: Emma BarnabyMusic composition: Eleni ShawSound Engineer: Nigel AppletonEditor: David PrestCommissioned by DeepMind Thank you to everyone who made this season possible!  Further reading: GPT-3 Powers the Next Generation of Apps, OpenAI: https://openai.com/blog/gpt-3-apps/https://web.stanford.edu/class/linguist238/p36-weizenabaum.pdfNever Mind the Computer 1983 about the ELIZA program, BBC: https://www.bbc.co.uk/programmes/p023kpf8How Large Language Models Will Transform Science, Society, and AI, Stanford University: https://hai.stanford.edu/news/how-large-language-models-will-transform-science-society-and-aiChallenges in Detoxifying Language Models, DeepMind: https://deepmind.com/research/publications/2021/Challenges-in-Detoxifying-Language-ModelsExtending Machine Language Models toward Human-Level Language Understanding, DeepMind: https://deepmind.com/research/publications/2020/Extending-Machine-Language-Models-toward-Human-Level-Language-UnderstandingLanguage modelling at scale, DeepMind: https://deepmind.com/blog/article/language-modelling-at-scaleArtificial general intelligence, Technology Review: https://www.technologyreview.com/2020/10/15/1010461/artificial-general-intelligence-robots-ai-agi-deepmind-google-openai/A Definition of Machine Intelligence by Shane Legg, arXiv: https://arxiv.org/abs/0712.3329Stuart Russell - Living With Artificial Intelligence, BBC: https://www.bbc.co.uk/programmes/m001216k/episodes/player Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

In December 2019, DeepMind’s AI system, AlphaFold, solved a 50-year-old grand challenge in biology, known as the protein-folding problem. A headline in the journal Nature read, “It will change everything” and the President of the UK's Royal Society called it a “stunning advance [that arrived] decades before many in the field would have predicted”. In this episode, Hannah uncovers the inside story of AlphaFold from the people who made it happen and finds out how it could help transform the future of healthcare and medicine.For questions or feedback on the series, message us on Twitter @DeepMind or email podcast@deepmind.com. Interviewees: DeepMind’s Demis Hassabis, John Jumper, Kathryn Tunyasunakool and Sasha Brown; Charles Mowbray and Monique Wasuna of the Drugs for Neglected Diseases initiative (DNDi]) & John McGeehan of the Centre for Enzyme Innovation at the University of Portsmouth CreditsPresenter: Hannah FrySeries Producer: Dan HardoonProduction support: Jill AchinekuSounds design: Emma BarnabyMusic composition: Eleni ShawSound Engineer: Nigel AppletonEditor: David PrestCommissioned by DeepMind Thank you to everyone who made this season possible!  Further reading: AlphaFold blog, DeepMind: https://deepmind.com/blog/article/alphafold-a-solution-to-a-50-year-old-grand-challenge-in-biologyAlphaFold case study, DeepMind: https://deepmind.com/research/case-studies/alphafoldIt will change everything, Nature: https://www.nature.com/articles/d41586-020-03348-4AlphaFold Is The Most Important Achievement In AI—Ever, Forbes: https://www.forbes.com/sites/robtoews/2021/10/03/alphafold-is-the-most-important-achievement-in-ai-ever/?sh=359278426e0aBacteria found to eat PET plastics, NewScientist: https://www.newscientist.com/article/2080279-bacteria-found-to-eat-pet-plastics-could-help-do-the-recycling/Protein Structure Prediction Center: https://predictioncenter.org/An interview with Professor John McGeehan, BBSRC: https://bbsrc.ukri.org/news/features/enzyme-science/an-interview-with-professor-john-mcgeehan/John McGeehan profile, University of Portsmouth: https://researchportal.port.ac.uk/en/persons/john-mcgeehanDrugs for Neglected Diseases initiative (DNDi): https://dndi.org/A doctor’s dream, DNDi: https://www.youtube.com/watch?v=Tk31iucWYdEThe Curious Cases of Rutherford and Fry, BBC: https://www.bbc.co.uk/programmes/b07dx75g/episodes/downloadsHannah Fry: https://hannahfry.co.uk/ Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

The chart-topping podcast which uncovers the extraordinary ways artificial intelligence (AI) is transforming our world is back for a second season. Join mathematician and broadcaster Professor Hannah Fry behind the scenes of world-leading AI research lab DeepMind to get the inside story of how AI is being created – and how it can benefit our lives and the society we live in.Recorded over six months and featuring over 30 original interviews, including DeepMind co-founders Demis Hassabis and Shane Legg, the podcast gives listeners exclusive access to the brilliant people building the technology of the future. Throughout nine original episodes, Hannah discovers how DeepMind is using AI to advance science in critical areas, like solving a 50-year-old grand challenge in biology and developing nuclear fusion.Listeners hear stories of teaching robots to walk at home during lockdown, as well as using AI to forecast weather, help people regain their voices, and enhance game strategies with Liverpool Football Club. Hannah also takes an in-depth look at the challenges and potential of building artificial general intelligence (AGI) and explores what it takes to ensure AI is built to benefit society.“I hope this series gives people a better understanding of AI and a feeling for just how exhilarating an endeavour it is.” – Demis Hassabis, CEO and Co-Founder of DeepMindFor questions or feedback on the series, message us on Twitter @DeepMind or email podcast@deepmind.com.CreditsPresenter: Hannah FrySeries Producer: Dan HardoonProduction support: Jill AchinekuSounds design: Emma BarnabyMusic composition: Eleni ShawSound Engineer: Nigel AppletonEditor: David PrestCommissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

In this special extended episode, Hannah Fry meets Demis Hassabis, the CEO and co-founder of DeepMind. She digs into his former life as a chess player, games designer and neuroscientist and explores how his love of chess helped him to get start-up funding, what drives him and his vision, and why AI keeps him up at night.If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or email us at podcast@deepmind.com.Further reading:Wired: Inside DeepMind's epic mission to solve science's trickiest problemQuanta magazine: How Artificial Intelligence Is Changing ScienceDemis Hassabis: A systems neuroscience approach to building AGI. Talk at the 2010 Singularity Summit Demis Hassabis: The power of self-learning systems. Talk at MIT 2019Demis Hassabis: Talk on Creativity and AI Financial Times: The mind in the machine: Demis Hassabis on artificial intelligence (2017)The Times: Interview with Demis HassabisThe Economist Babbage podcast: DeepMind GamesInterview with Demis Hassabis from the book Game Changer, which also features an introduction from DemisInterviewees: Deepmind CEO and co-founder, Demis HassabisCredits:Presenter: Hannah FryEditor: David PrestSenior Producer: Louisa FieldProducers: Amy Racs, Dan HardoonBinaural Sound: Lucinda Mason-BrownMusic composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)Commissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

AI researchers around the world are trying to create a general purpose learning system that can learn to solve a broad range of problems without being taught how. Koray Kavukcuoglu, DeepMind’s Director of Research, describes the journey to get there, and takes Hannah on a whistle-stop tour of DeepMind’s HQ and its research.If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or email us at podcast@deepmind.com.Further reading:OpenAI: An overview of neural networks and the progress that has been made in AIShane Legg, DeepMind co-founder: Measuring machine intelligence at the 2010 Singularity SummitShane Legg and Marcus Hutter: Paper on defining machine intelligenceDemis Hassabis: Talk on the history, frontiers and capabilities of AIRobert Wiblin: Positively shaping the development of artificial intelligenceAsilomar AI PrinciplesRichard S. Sutton and Andrew G. Barto: Reinforcement Learning: An IntroductionInterviewees: Koray Kavukcuoglu, Director of Research; Trevor Back, Product Manager for DeepMind’s science research; research scientists Raia Hadsell and Murray Shanahan; and DeepMind CEO and co-founder, Demis Hassabis.Credits:Presenter: Hannah FryEditor: David PrestSenior Producer: Louisa FieldProducers: Amy Racs, Dan HardoonBinaural Sound: Lucinda Mason-BrownMusic composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)Commissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

While there is a lot of excitement about AI research, there are also concerns about the way it might be implemented, used and abused. In this episode Hannah investigates the more human side of the technology, some ethical issues around how it is developed and used, and the efforts to create a future of AI that works for everyone.If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or email us at podcast@deepmind.com.Further reading:The Partnership on AIProPublica: investigation into machine bias in criminal sentencingScience Museum – free exhibition: Driverless: who is in control (until Oct 2020)Survival of the best fit: An interactive game that demonstrates some of the ways in which bias can be introduced into AI systems, in this case for hiringJoy Buolamwini: AI, Ain’t I a Woman: A spoken word piece exploring AI bias, and systems not recognising prominent black womenHannah Fry: Hello World - How to be Human in the Age of the MachineDeepMind: Safety and EthicsFuture of Humanity Institute: AI Governance:A Research AgendaInterviewees: Verity Harding, Co-Lead of DeepMind Ethics and Society; DeepMind’s COO Lila Ibrahim, and research scientists William Isaac and Silvia Chiappa.Credits:Presenter: Hannah FryEditor: David PrestSenior Producer: Louisa FieldProducers: Amy Racs, Dan HardoonBinaural Sound: Lucinda Mason-BrownMusic composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)Commissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

The ambition of much of AI research is to create systems that can help to solve problems in the real world. In this episode, Hannah meets the people building systems that could be used to save the sight of thousands, help us solve one of the most fundamental problems in biology and reduce energy consumption in an effort to combat climate change. But whilst there is great potential, there are also important obstacles that will need to be tackled for AI to be used effectively, safely and fairly.If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or email us at podcast@deepmind.com.Further reading:Wired: Inside DeepMind's epic mission to solve science's trickiest problemDeepMind blogs on the partnership with Moorfields NHS eye hospital and predicting eye disease, and Moorfields’ news announcement on its research with DeepMindDeepMind blog: AlphaFold: Using AI for scientific discoveryDeepMind blogs on reducing Google’s energy bill for datacentre cooling and how this project has progressedResearch paper: Tackling Climate Change with Machine LearningQuanta magazine: How Artificial Intelligence Is Changing ScienceDeepMind blog: How evolutionary selection can train more capable self-driving carsOther examples of the application of AI for real-world impact include:Francis Crick Institute: machine learning models that can help predict heart diseaseNASA: AUDREY machine learning system to better guide first responders through firesUniversity of Southern California: Protection Assistant for Wildlife Security using AI to help wildlife conservationInterviewees: Pearse Keane, consultant ophthalmologist at Moorfields Eye Hospital; Sandy Nelson, Product Manager for DeepMind’s Science Program; and DeepMind Program Manager Sims Witherspoon.Credits:Presenter: Hannah FryEditor: David PrestSenior Producer: Louisa FieldProducers: Amy Racs, Dan HardoonBinaural Sound: Lucinda Mason-BrownMusic composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)Commissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

Forget what sci-fi has told you about superintelligent robots that are uncannily human-like; the reality is more prosaic. Inside DeepMind’s robotics laboratory, Hannah explores what researchers call ‘embodied AI’: robot arms that are learning tasks like picking up plastic bricks, which humans find comparatively easy. Discover the cutting-edge challenges of bringing AI and robotics together, and learning from scratch how to perform tasks. She also explores some of the key questions about using AI safely in the real world.If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or email us at podcast@deepmind.com.Further reading:Blogs on AI safety and further resources from Victoria KrakovnaThe Future of Life Institute: The risks and benefits of AIThe Wall Street Journal: Protecting Against AI’s Existential ThreatTED Talks: Max Tegmark - How to get empowered, not overpowered, by AIRoyal Society lecture series sponsored by DeepMind: You & AINick Bostrom: Superintelligence: Paths, Dangers and Strategies (book)OpenAI: Learning from Human PreferencesDeepMind blog: Learning from human preferencesDeepMind blog: Learning by playing - how robots can tidy up after themselvesDeepMind blog: AI safetyInterviewees: Software engineer Jackie Kay and research scientists Murray Shanahan, Victoria Krakovna, Raia Hadsell and Jan Leike.Credits:Presenter: Hannah FryEditor: David PrestSenior Producer: Louisa FieldProducers: Amy Racs, Dan HardoonBinaural Sound: Lucinda Mason-BrownMusic composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)Commissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.  

Video games have become a favourite tool for AI researchers to test the abilities of their systems. In this episode, Hannah sits down to play StarCraft II - a challenging video game that requires players to control the onscreen action with as many as 800 clicks a minute. She is guided by Oriol Vinyals, an ex-professional StarCraft player and research scientist at DeepMind, who explains how the program AlphaStar learnt to play the game and beat a top professional player. Elsewhere, she explores systems that are learning to cooperate in a digital version of the playground favourite ‘Capture the Flag’.If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or emailing us at podcast@deepmind.com.Further readingThe Economist: Why AI researchers like video gamesDeepMind blogs: Capture the Flag and AlphastarProfessional StarCraft II player MaNa gives his impressions of AlphaStar and DeepMindOpen AI’s work on Dota 2 The New York Times: DeepMind can now beat us at multiplayer games, tooRoyal Society: Machine Learning resourcesDeepMind: The Inside Story of AlphaStar Andrej Karpathy: Deep Reinforcement Learning: Pong from PixelsInterviewees: Research scientists Max Jaderberg and Raia Hadsell; Lead researchers David Silver and Oriol Vinyals, and Director of Research Koray Kavukcuoglu.Credits:Presenter: Hannah FryEditor: David PrestSenior Producer: Louisa FieldProducers: Amy Racs, Dan HardoonBinaural Sound: Lucinda Mason-BrownMusic composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)Commissioned by DeepMind Please like and subscribe on your preferred podcast platform. Want to share feedback? Or have a suggestion for a guest that we should have on next? Leave us a comment on YouTube and stay tuned for future episodes.