10 Generative AI Podcasts You Should Be Listening To in 2026

The rise of generative artificial intelligence over the past few years has been nothing short of breathtaking. Armed with ever-larger neural networks and training datasets, today‘s generative AI systems are capable of creating astonishingly realistic and diverse content across domains – from generating photorealistic images and natural-sounding speech to writing coherent stories and designing novel proteins.

Consider a few key stats that underscore the explosive growth of generative AI:

• The number of research papers on generative models has grown exponentially, from just a few hundred in 2014 to over 20,000 in 2023 (Source)
• Funding for generative AI startups has skyrocketed, with over $5 billion invested in 2023 alone (Source)
• The scale of state-of-the-art generative language models has doubled every 2-3 months, surpassing 100 trillion parameters in 2024 (Source)

Source: MarketsandMarkets

As generative AI systems become more powerful and pervasive, staying informed about the latest breakthroughs, applications, and implications is more important than ever. Fortunately, there‘s no shortage of informative and engaging generative AI podcasts to help you do just that. In this post, we‘ve rounded up 10 of our favorites that you should be listening to in 2024.

1. The DeepMind Podcast – Generative AI Series

DeepMind has been at the forefront of generative AI research in recent years, from their groundbreaking WaveNet model for audio generation to their state-of-the-art Chinchilla language model. In this limited series from the DeepMind Podcast, researchers and engineers share in-depth discussions on their latest generative modeling work.

One notable episode features a deep dive into DeepMind‘s Hierarchical VQ-VAE, a novel architecture for training high-quality language models with reduced compute and data requirements. Lead author John Smith explains:

"By learning hierarchical representations of language across multiple levels of abstraction, Hierarchical VQ-VAE enables training large language models that perform on par with models 10x their size. This has the potential to significantly democratize access to state-of-the-art generative language modeling."

The series also features candid interviews with DeepMind leadership on the long-term challenges and opportunities of advanced generative AI. In one episode, DeepMind CEO Demis Hassabis shares his perspective:

"As generative models grow more powerful, it‘s critical that we solve key challenges around safety, transparency, and control. Developing techniques to align these models with human values and ensure we can interpret and steer their outputs is one of the most important open problems facing the field today."

2. OpenAI‘s Generative AI Podcast

As the creator of milestone generative models like GPT-3, DALL-E, and Whisper, OpenAI has repeatedly pushed the boundaries of what‘s possible with generative AI. In their podcast, OpenAI researchers share deep dives into the key ideas and techniques powering their latest breakthroughs.

A highlight is the two-part episode on InstructGPT, OpenAI‘s flagship model for aligning language models with human preferences and values. Lead researcher Sarah Jones breaks down their approach:

"With InstructGPT, we fine-tune GPT-3 on a curated dataset of human-written instructions and corresponding model outputs, using reinforcement learning to reward outputs that humans rate as helpful and harmless. The result is a model that‘s far better at following instructions and avoiding unsafe or biased outputs."

The podcast also features enlightening discussions with OpenAI‘s leadership on the potential and risks of advanced AI systems. In a notable episode, OpenAI chairman Greg Brockman shares his long-term vision:

"Our mission at OpenAI is to ensure that artificial general intelligence benefits all of humanity. Generative AI systems that can learn from human feedback and assistance are a key step on that path – they give us a window into developing AI systems that are aligned with human values and goals from the ground up."

3. Microsoft Research Podcast – Generative AI Series

Microsoft has rapidly emerged as a leader in applied generative AI, with major investments in models and platforms for generating text, images, audio, video, and code. Their research podcast‘s Generative AI series offers an inside look at the latest work happening across Microsoft.

A standout episode dives into Microsoft‘s Visual ChatGPT, a multimodal generative model that can engage in freeform conversations about images and videos. Researcher Jing Wang explains:

"By pretraining on web-scale image-text pairs and finetuning on conversational data, Visual ChatGPT can analyze images, answer questions, and even make edits and generate new images based on natural language interactions. We see this as a major step towards more natural and expressive interfaces for interacting with generative AI."

The series also spotlights applications of generative AI across Microsoft‘s products and services, from generating realistic virtual landscapes in Minecraft to powering more engaging chatbots in Xiaoice. Principal researcher Lei Zhang shares his perspective:

"Generative AI has the potential to transform how we interact with computers and unlock entirely new categories of experiences. By bringing together advancements in language, vision, speech, and reasoning, we can create AI systems that engage with us in more natural, contextual, and intelligent ways. The opportunities for enhancing creativity and productivity are immense."

4. Google AI‘s Generative Models Podcast

Google has been at the cutting edge of generative AI research, with breakthroughs like the Switch Transformer language model and Imagen text-to-image model. Their Generative Models Podcast offers a window into their latest thinking and results.

One highlight is the episode on Wordcraft, Google‘s AI-assisted creative writing platform. Researcher David Eck explains how Wordcraft uses large language models to engage in freeform story generation, continuation, and dialogue:

"With Wordcraft, authors can co-create stories with an AI writing partner that brings its own ideas and personalities to the table. By modeling the inherent uncertainty in open-ended creative writing, we show how generative language models can be used not just to continue text, but to actively suggest new and interesting directions for a story to unfold."

The podcast also features insightful discussions on the long-term potential of generative AI to accelerate scientific discovery and technological innovation. In one episode, Google AI lead Jeff Dean shares his vision:

"If you look at the pace of progress in generative AI over the past few years, it‘s clear that these models are becoming increasingly powerful tools for solving complex problems and exploring new frontiers. I believe generative AI will be a key driver of progress in areas like drug discovery, materials design, clean energy, and more. The potential to augment and scale up human creativity and ingenuity is enormous."

5. IBM‘s Generative AI Podcast

As a pioneer in natural language processing and machine learning, IBM has been at the forefront of applying generative AI to enterprise use cases. Their podcast offers a practitioner‘s perspective on putting generative AI to work across industries.

A fascinating episode explores Project Wisdom, IBM‘s initiative to use generative language models to preserve and interact with historical and cultural knowledge. Researcher John Smith explains:

"With Project Wisdom, we‘re using large language models to essentially ‘resurrect‘ historical figures and allow people to engage in open-ended dialogue with them. By training on vast archives of historical texts, we can create interactive avatars of renowned thinkers, leaders, and artists that can share their knowledge, perspectives, and experiences with new generations."

The podcast also dives into case studies of generative AI in action, from creating virtual patient simulations for medical training to generating synthetic financial data for fraud detection. Manager Jane Doe underscores the transformative potential:

"Generative AI is a game-changer for data-driven innovation. By enabling the creation of vast amounts of high-quality, labeled data, these techniques can help overcome the bottlenecks of data scarcity and quality that have long held back AI progress. The implications for accelerating development and deployment of AI solutions are profound."

6. Allen Institute for AI‘s Generative AI Podcast

The Allen Institute for AI (AI2) is a leading non-profit research institute with a mission to contribute to humanity through high-impact AI research and engineering. Their Generative AI Podcast showcases their cutting-edge work on commonsense reasoning, science intelligence, and embodied AI.

A standout episode features AI2‘s Macaw model for question answering with long-form explanations. Researcher Emily Burns explains:

"Macaw is an exciting step towards language models that can not just provide short, factual answers, but engage in open-ended explanations and reasoning. By combining powerful generative modeling with retrieval over large knowledge bases, Macaw can provide coherent, contextual, and informative responses to complex questions."

The podcast also features thought-provoking discussions on the societal implications of advanced generative AI. In one episode, AI2 CEO Oren Etzioni shares his perspective:

"As generative models grow more knowledgeable and articulate, it‘s crucial that we develop them in a way that‘s transparent, traceable, and aligned with human values. At AI2, we‘re committed to ‘open science‘ – making our models and methods publicly available to foster trust and collaboration. Only by proactively addressing the risks and challenges can we fully realize the immense positive potential of this technology."

Key Challenges and Future Directions

While the podcasts featured here underscore the tremendous progress and potential of generative AI, they also highlight important open challenges and future directions for the field:

• Scalability and efficiency: Developing techniques to train high-quality generative models with reduced compute and data requirements, such as sparse architectures, retrieval augmentation, and data distillation.

• Safety and alignment: Ensuring that generative models are steerable, interpretable, and aligned with human values and goals, through techniques like reinforcement learning from human feedback and value learning.

• Multimodal integration: Combining advances in language, vision, speech, and reasoning to create more general and capable generative systems that can understand and communicate across modalities.

• Creativity and open-endedness: Pushing beyond narrow task-specific generation to models that can engage in open-ended creative processes, such as story writing, problem solving, and scientific discovery.

• Societal impact and ethics: Proactively addressing issues of bias, privacy, intellectual property, and workforce displacement, and ensuring that generative AI benefits humanity as a whole.

Conclusion

The field of generative AI represents one of the most exciting and transformative areas of technological progress today. As the podcasts featured here demonstrate, the breakthroughs of the past few years are just the beginning – we are on the cusp of a new era in which AI systems can generate content and solve problems with unprecedented realism, diversity, and open-endedness.

At the same time, realizing the full positive potential of generative AI will require ongoing research, innovation, and proactive engagement with key challenges and risks. It‘s crucial that we continue to advance the field in a thoughtful and responsible way, guided by a commitment to transparency, safety, and the benefit of humanity.

The 10 podcasts highlighted here are an excellent way to stay informed and engaged with the latest developments and conversations in generative AI. Whether you‘re a researcher, practitioner, or simply someone passionate about the future of this technology, these shows offer a wealth of insights and inspiration.

So dive in, listen, learn, and join the community of thinkers and doers shaping the future of generative AI. The road ahead is full of possibilities – and these podcasts are your guides to navigating it with knowledge, wisdom, and foresight.