Week 9 of Podcast Recommendations: AI

I’m in Week 9 of 12 for podcast recommendations: this week is all about artificial intelligence. Mix and match these episodes to create your own at-home AI “conference.”?There are a number of prominent AI academics and researchers on this list. And big thanks to the "podcasts" subreddit community which turned out to be a fantastic resource when searching for interesting, accessible AI podcasts.

AI Fundamentals

If you or anyone you know wants to understand what AI is, how it works and why generative AI and ChatGPT are in the news every day, this is a good place to start:

“Raphael Milliere on How Artificial Intelligence Thinks.”? Sean Carroll’s Mindscape. ??Raphael is a philosophy professor at Columbia University; Sean is a Johns Hopkins professor who hosts insightful interviews with the smartest people in the world on topics much broader than AI. This is a good primer on gen AI and large language models. (1 hour 57 minutes)

“The Biggest Event in Human History.” The Reith Lectures. ? These annual lectures from the BBC have a history that dates back decades and cover important issues of the day.?The 2021 series was an excellent group of 4 lectures (Lecture 2 , Lecture 3 and Lecture 4 ) about AI, presented by Berkeley computer science professor Stuart Russell in a timely and timeless manner.?Don’t miss the Q&A at the end of each session; there are cameos from luminaries like Tim Berners-Lee. ??(~1 hour per episode, 4 episodes)

“Munk Dialogue with AI Debaters Yann Lecun, Max Tegmark, Melanie Mitchell and Yoshua Bengio.” The Munk Debates Podcast. ?This Canadian series ponders many of the “significant issues of our time.”?This caught my eye because the Chief AI Scientist at Google was one of the speakers as well as three other respected AI researchers. (1 hour)

“Computer Scientist Explains Machine Learning in 5 Levels of Difficulty.” The clickbait title aside, this is a useful YouTube video, with people of all ages learning about and explaining machine learning and AI, using elementary to complex language and terms. (26 minutes)

“The AI Revolution Could Be Bigger and Weirder Than We Can Imagine.” Plain English with Derek Thompson. ?This episode is from a few months ago but it’s perhaps the most easily digestible summary of the podcasts above. (1 hour)?

Bear Arguments

There’s a lot that we’ve already heard about AI bias and misinformation (though if you want more on that, just look for anything from Emily Bender, Timnit Gebru or Meredith Broussard).?Here are other important issues that are less discussed but major pitfalls of AI:

“AI Is Coming for Your Children.” Behind the Bastards. ?It turns out that many of the dinosaur coloring books on Amazon were created by generative AI through Kindle Direct Publishing.?Not surprisingly, they are terrible in many ways. ?This podcast makes a compelling case for how bad a slippery slope we’re on when we let lazy, greedy and/or nefarious people use gen AI to create content for others, especially kids. (1 hour 16 minutes)

“An AI Chatbot Debate – with Blake Lemoine and Gary Marcus.” Big Technology Podcast. ?Lemoine became famous after being fired by Google shortly after claiming the company’s LaMDA chatbot was sentient.?Marcus is one of the most outspoken critics of generative AI these days. ?The first part of the episode is a debate about whether AI is really sentient; the second part has both speakers agreeing that regulation, guardrails and transparency around AI are necessary. (1 hour)

“Deepfake Law 101.” ?Endless Thread. ?This is mostly about Section 230 (which I still maintain is the single worst piece of legislation created in our lifetime, thank you Chris Cox and Ron Wyden) and how it may impede the creation of any law about deepfakes, which just about everyone thinks is necessary.?(33 minutes)

“Hearing on the Use of AI in the Military and Combat.”? This C-Span recording of a recent Congressional hearing was frightening because it talks about the necessary regulatory framework crucial for AI, none of which the US or the world has in place. ?As in so many Congressional hearings, the naivety of politicians about crucial topics like Chinese politics or when dazzled by Silicon Valley showmen is on prominent display. (1 hour 33 minutes)

“Bot Love.” Radiotopia. This 2023 series interviews people who have found comfort, camaraderie and companionship with bots.?It is a real-life version of the dystopian movie Her, and highlights the particular appeal that bots have for vulnerable individuals with mental health issues.?(~24 minutes per episode, 7 episodes)

“The Black Box: In AI We Trust?” Vox Unexplainable. ?This episode pulls together much of the concern we all have about mysterious AI algorithms and their unintended consequences. (30 minutes)

Bull Arguments

The bear arguments aside, AI tools, especially tightly controlled machine learning applications, have been positively impacting many industries for years. Here are some contemporary examples of specific sector use cases:

“Making 600 Billion Decisions with AI: Expedia Group’s Rathi Murthy” and “Partnerships in AI Drive Conservation Efforts: WWF’s Dave Thau.” Me, Myself and AI. This series interviews diverse practitioners about data science.?Two recent podcasts featured the CTO of Expedia, which provides a great parallel for the use of data in eCommerce, and the World Wildlife Fund which has a number of interesting AI use cases in conservation that we rarely hear about. (33 and 26 minutes)

“MIT’s Anant Agarwal on AI in Education.” AI Podcast by Nvidia .?Anant is the founder of EdX which is one of the world’s large MOOC platforms.?This is an enthusiastic summary of how personalized education and AI-generated problem sets can empower students and teachers around the world. There are parallels to training in business as well. ?(38 minutes)

“Lee Hood & Nathan Price.” Armchair Expert . ?Dax Shepherd’s podcast is a popular title on Spotify and this episode is a long one that dives into the applications of AI in personalized medicine and drug creation. The promise of AI on disease treatment is perhaps one of the most anticipated positive impacts.?(2 hours 13 minutes)

“AI Could Solve Some of Humanity’s Hardest Problems. It Already Has.” The Ezra Klein Show. ?I thought I could go 12 weeks without making an Ezra Klein recommendation but I could not.?The head of AlphaFold, a product of Google’s DeepMind, which identifies proteins, gives a fascinating interview about the future of biology and drug discovery. (1 hour 28 minutes)

Using AI. ? This podcast series of 10 episodes created by AI industry practitioners discusses a range of gen AI use cases from news to marketing to legal documents to code creation.?(~50 minutes per episode, 10 episides)

“Rodrigo Liang, SambaNova CEO, shares how he raised more than $1B to launch and grow the first generative AI unicorn.” AI and the Future of Work.? This podcast interviews founders of a number of AI startups.?This particular conversation is with the founder of SambaNova, an AI unicorn that promises to protect company data while giving everyone in the world their own gen AI personal assistant.?(42 minutes)

“Automated cartography using AI.” Practical AI. This podcast is hosted by an AI practitioner; this episode is about government uses of open source data for civic goals.?(44 minutes)

?