When we started this Podcast back in August 2022, we, Calum and David, announced the theme to be “Anticipating and managing exponential impact”. We talked about three sub-themes: Developing the skills of exponential foresight; Distinguishing between scenarios, whether they were plausible or implausible, and whether they were desirable or undesirable; and thirdly, Supporting the community of collaborative exponential foresight. 126 episodes later, as we reach the transition between 2025 and 2026, it’s a good time for the two of us to take stock.

Accordingly, in this episode, we each pick out a number of events from the last 12 months which we see as potential signals of larger exponential impact ahead.

Selected follow-ups:

• An MIT report that 95% of AI pilots fail spooked investors - by Jeremy Kahn
• The Shape of AI: Jaggedness, Bottlenecks and Salients - by Ethan Mollick
• The Road To Superintelligence - by Calum
• AI Doomers, Accelerationists & Scouts - Digital Disruption
• The Economic Singularity - book by Calum
• How can better foresight actually improve the world? - Webinar in the series "From forecasts to levers"
• Disrupting the first reported AI-orchestrated cyber espionage campaign - Anthropic
• Major Neuromorphic Computing projects - listed by Conscium
• Why AI Agent Verification Is A Critical Industry - by Calum
• Climate change and populism: Grounds for optimism? - LFP episode with Matt Burgess
• What's Our Problem? - book by Tim Urban
• OpenAI and Retro Biosciences achieve 50x increase in expressing stem cell reprogramming markers
• Progress at LEVF, December 2025 - by David
• UK Biobank

Listen on: Apple Podcasts Spotify

Listen on: Apple Podcasts Spotify

Almost every serious discussion about options to constrain the development of advanced AI results in someone raising the question: “But what about China?” The worry behind this question is that slowing down AI research and development in the US and Europe will allow China to race ahead.

It's true: the relationship between China and the rest of the world has many complications. That’s why we’re delighted that our guest in this episode is Kayla Blomquist, the Co-founder and Director of the Oxford China Policy Lab, or OCPL for short. OCPL describes itself as a global community of China and emerging technology researchers at Oxford, who produce policy-relevant research to navigate risks in the US-China relationship and beyond.

In parallel with her role at OCPL, Kayla is pursuing a DPhil at the Oxford Internet Institute. She is a recent fellow at the Centre for Governance of AI, and the lead researcher and contributing author to the Oxford China Briefing Book. She holds an MSc from the Oxford Internet Institute and a BA with Honours in International Relations, Public Policy, and Mandarin Chinese from the University of Denver. She also studied at Peking University and is professionally fluent in Mandarin.

Kayla previously worked as a diplomat in the U.S. Mission to China, where she specialized in the governance of emerging technologies, human rights, and improving the use of new technology within government services.

Selected follow-ups:

• Kayla Blomquist - Personal site
• Oxford China Policy Lab
• The Oxford Internet Institute (OII)
• Google AI defeats human Go champion (Ke Jie)
• AI Safety Summit 2023 (Bletchley Park, UK)
• United Kingdom: Balancing Safety, Security, and Growth - OCPL
• China wants to lead the world on AI regulation - report from APEC 2025
• China's WAICO proposal and the reordering of global AI governance
• Impact of AI on cyber threat from now to 2027
• Options for the future of the global governance of AI - London Futurists Webinar
• A Tentative Draft of a Treaty - Online appendix to the book If Anyone Builds It, Everyone Dies
• An International Agreement to Prevent the Premature Creation of Artificial Superintelligence

Music: Spike Protein, by Koi Discovery, available under CC0 1.0 Public Domain Declaration

Listen on: Apple Podcasts Spotify

Our guest in this episode is Stephen Witt, an American journalist and author who writes about the people driving the technological revolutions. He is a regular contributor to The New Yorker, and is famous for deep-dive investigations.

Stephen's new book is "The Thinking Machine: Jensen Huang, Nvidia, and the World's Most Coveted Microchip", which has just won the 2025 Financial Times and Schroders Business Book of the Year Award. It is a definitive account of the rise of Nvidia, from its foundation in a Denny's restaurant in 1993 as a video game component manufacturer, to becoming the world's most valuable company, and the hardware provider for the current AI boom.

Stephen's previous book, “How Music Got Free”, is a history of music piracy and the MP3, and was also a finalist for the FT Business Book of the Year.

Selected follow-ups:

• Stephen Witt - personal site
• Articles by Stephen Witt on The New Yorker
• The Thinking Machine: Jensen Huang, Nvidia, and the World's Most Coveted Microchip - book site
• Stephen Witt wins FT and Schroders Business Book of the Year - Financial Times
• Nvidia Executives
• Battle Royale (Japanese film) - IMDb
• The Economic Singularity - book by Calum Chace
• A Cubic Millimeter of a Human Brain Has Been Mapped in Spectacular Detail - Nature
• NotebookLM - by Google

Music: Spike Protein, by Koi Discovery, available under CC0 1.0 Public Domain Declaration

Listen on: Apple Podcasts Spotify

Listen on: Apple Podcasts Spotify