The method of invention, AI's new clock speed and why capital markets are confused

Welcome to Exponential View, the show where I explore how exponential technologies such as AI are reshaping our future. I've been studying AI and exponential technologies at the frontier for over ten years. Each week, I share some of my analysis or speak with an expert guest to make light of a particular topic. To keep up with the Exponential transition, subscribe to this channel or to my newsletter: https://www.exponentialview.co/ --- In this episode, I look at the next 24 months of AI. The technology is improving rapidly – so what could hold back widespread transformation of how we work and live? I dig into the real constraints, from electricity shortages to institutional inertia, why mid-2026 matters for enterprise AI, and why so many people remain uneasy about a technology they use every day. I cover: (00:03) Predicting AI's next two years (01:50) How life changing are chatbots, really? (03:36) Our current biggest AI constraint (07:58) The remarkable increase in token efficiency (10:43) Why mid-2026 is a crucial turning point (13:01) Do we actually want AI in our lives? (15:28) Should organizations wait to jump in? (16:39) How is OpenAI reckoning with Gemini? (18:41) The market's reaction to OpenAI's code red (19:32) Where will value accrue in the supply chain? (20:51) What's the best strategy for middling powers?Where to find me: Exponential View newsletter: https://www.exponentialview.co/ Website: https://www.azeemazhar.com/ LinkedIn: https://www.linkedin.com/in/azhar/Twitter/X: https://x.com/azeem Production by supermix.io and EPIIPLUS1 Production and research: Chantal Smith and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Welcome to Exponential View, the show where I explore how exponential technologies such as AI are reshaping our future. I've been studying AI and exponential technologies at the frontier for over ten years. Each week, I share some of my analysis or speak with an expert guest to make light of a particular topic. To keep up with the Exponential transition, subscribe to this podcast or to my newsletter: https://www.exponentialview.co/ --- In this episode, I reflect on the third anniversary of ChatGPT's launch as a marker of where we are in the exponential age. As a product, ChatGPT captures the speed of technological progress, the new behaviours emerging around it and the widening gap between innovation and institutional change – all symptomatic of the era I called the exponential age in my 2021 book. I cover: (00:09) How ChatGPT became synonymous with AI (01:41) The rise of the reasoning model (03:53) Why NVIDIA's chip cycle is exponential (05:53) How general-purpose tech changes everything (07:59) The subtle power of building bespoke software (11:46) The iPhone calculation that breaks everything (14:53) Who profits from a general-purpose technology? (16:38) The software market example (20:07) Are we headed towards another .com crash?  Where to find me: Exponential View newsletter: https://www.exponentialview.co/ Website: https://www.azeemazhar.com/ LinkedIn: /azhar Twitter/X: https://x.com/azeem Production by supermix.io and EPIIPLUS1 Production and research: Chantal Smith and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

The AI industry is sending mixed signals, with markets turning red while teams report real productivity gains. In this session I explore why we are living in a split reality, where individuals move faster with these tools but the wider economy is ambivalent. We once assumed juniors would get the biggest lift from AI, yet the newer agentic tools seem to reward senior workers who know how to structure problems and judge output. In this podcast, I look at the evidence behind that shift and explain how these gains collide with the slow grind of organisational processes. I cover: (00:00) AI productivity: A split reality (00:31) Decoding the stock market drop (02:53) Unpacking three years of AI productivity data (06:09) Does AI help junior or senior developers more? (09:54) The surprising group benefitting from AI (11:45) Why is there a productivity gap? (13:08) Most companies need a process overhaul (14:33) Anthropic's alarming discovery (16:45) So, are we moving quickly enough? (17:29) The counterintuitive truth about AI productivity  Where to find me:- Exponential View newsletter: https://www.exponentialview.co/ - Website: https://www.azeemazhar.com/ - - LinkedIn: https://www.linkedin.com/in/azhar - Twitter/X: https://x.com/azeem  Production by supermix.io and EPIIPLUS1 Production and research: Chantal Smith, Hannah Petrovic and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Junior roles in AI-exposed fields are disappearing fast. The obvious culprit is AI rapidly automating entry-level jobs. And yet, this isn't quite right. What is driving the drop is managers’ expectations about what AI will do, not the work that it's already replacing. I discussed this with Ben Zweig of Revelio Labs, which builds global workforce data from millions of individual profiles to track hiring, separations and job flows. Their data shows how expectation and uncertainty are reshaping the market.Together, we explored the future of work and shared practical advice for new grads. We covered: (01:15) What's happening in the labor market? (05:27) The inherent complexity of the labor market (06:24) How Revelio Labs captures labor market data (08:39) "The Canary in the Coal Mine" (11:52) Who does AI exposure harm the most? (13:01) How AI anticipation is harming the job market (15:15) Testing the expectation mismatch hypothesis (17:30) Could AI be creating more jobs? (20:44) Breaking down jobs into smaller tasks (27:33) Why large companies struggle to reorganize (30:35) Focus on creating adaptive, flexible roles (36:03) Managing AI's increasing capability (39:11) What entry-level workers need to do Where to find me: - Exponential View newsletter: https://www.exponentialview.co/ - Website: https://www.azeemazhar.com/ - LinkedIn: https://www.linkedin.com/in/azhar - Twitter/X: https://x.com/azeem Where to find Ben: - https://www.linkedin.com/in/ben-zweig/ - Twitter/X: https://x.com/BJZweig - Revelio Labs: https://www.reveliolabs.com/ Production by supermix.io and EPIIPLUS1 Production and research: Chantal Smith, Hannah Petrovic and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

The AI boom isn’t just about chatbots.In this video, I explain why cloud companies and chipmakers are exploding in value: we’re moving into an economy where computation becomes a fundamental input – like steel, electricity or oil.If that’s true, our demand for compute could approach infinity.I also break down new data from Wharton’s 2025 AI Adoption Report, which shows how AI agents and automated workflows are already spreading through major U.S. companies: https://knowledge.wharton.upenn.edu/special-report/2025-ai-adoption-report/Timestamps:(00:00) The economic shift to computation (00:40) The surprising Cloud business boom (02:52) Is the hardware industry growth a bubble? (03:18) What is computing, really? (04:31) Our insatiable appetite for computing (09:15) Our economic dependence on computation (10:54) The rise of agentic workforces (13:05) What does infinite demand actually mean? (15:23) The future of compute demandWhere to find me:Exponential View newsletter: https://www.exponentialview.co/Website: https://www.azeemazhar.com/LinkedIn: https://www.linkedin.com/in/azharTwitter/X: https://x.com/azeemProduction by supermix.io and EPIIPLUS1Production and research: Chantal Smith, Hannah Petrovic and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

In this episode, I speak with Jordan Schneider, creator of Chinatalk, to explore the new phase of US–China competition. Both countries are using trade policy, export controls and industrial strategy to shift the balance of global power. Yet, their economies remain tightly bound. We cover: (01:34) The US and China’s decoupling (07:28) Why attempts to control China backfired (08:51) Understanding the Oct. 9th rare Earth rules (11:27) The modern iteration of Chinese communism (14:23) Is decoupling a strategy to avoid weaponization? (16:12) US leadership might be shooting from the hip (19:22) Are system changes inherently messy? (21:27) “Vibe-based” sovereignty (26:03) AI incumbents aren’t entrenched—yet (29:07) Why China remains focused on AI deployment (32:45) The different versions of tech-accelerationism (33:37) How will societies withstand rapid change? (36:54) What the West can learn from China (40:10) Where China is most misunderstood (43:14) Imagining an improved US-China relationship Where to find Jordan:Substack: https://substack.com/@chinatalkYouTube: @ChinaTalkMedia‬Linkedin: / jorschneiderX: https://x.com/jordanschnycWhere to find me:Substack: https://www.exponentialview.co/Website: https://www.azeemazhar.com/LinkedIn:/ azharX: https://x.com/azeemProduced by EPIIPLUS1 Ltd and supermix.io Production and research: Chantal Smith, Hannah Petrovic and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Azeem Azhar sat down with Matthew Prince, co-founder & CEO of Cloudflare. Matthew is a rare operator with the vantage point to answer a simple question: if agents do the reading, who gets paid? This conversation is a practical map of how AI “answer engines” upend the web’s traffic-funded model – and what could replace it.Chapters: (00:46) The currency of the web is dying (06:08) Google's inflection point (10:08) Why a broken business model might save the internet (14:44) The incentivization of ragebait (20:38) Content scarcity as a solution (24:35) What could a new content business model look like? (28:51) The challenge of pricing information (29:31) How Cloudflare thinks about the creator economy (32:06) Should smaller companies pay less? (34:24) Can markets solve this without Congress? (39:11) How does the agentic web affect content? (43:40) A rare chance to redesign the internet Produced by EPIIPLUS1 Ltd and supermix.io Production and research: Chantal Smith, Hannah Petrovic, Nathan Warren and Marija Gavrilov. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

In this episode, I spoke with Dan Wang, author of “Breakneck: China’s Quest to Engineer the Future”, shortlisted for the FT & Schroders Business Book of the Year.Dan is one of the most astute observers of China’s technological and industrial development, and his annual letters from Beijing have long been required reading for those seeking to understand the country’s evolving role in the world.We unpacked a bold thesis: China is not merely a competitor in AI and tech, but is re-imagining its entire state apparatus as an engineering state - in contrast to the more “lawyerly” institutions of the US and UK.If you’re interested in AI, energy or geopolitics, this conversation is for you.We covered: (00:47) Why China is an engineering state(03:40) China’s pro-engineering disposition(06:08) The role of market competition in China(08:07) Living through Zero COVID(11:35) What political science terms get wrong(12:58) Characteristics of a lawyerly society(15:23) What Americans misunderstand about China(21:54) Has China produced essential tech?(23:50) The AI divide: China vs. US(27:45) Differences in energy production(32:07) The inherent value of process knowledge(38:34) Is the US developing pro-engineering policies?(44:23) What does it take for countries to compete?Where to find me:Substack: https://www.exponentialview.co/Website: https://www.azeemazhar.com/LinkedIn: https://www.linkedin.com/in/azharTwitter/X: https://x.com/azeemWhere to find Dan:Website: https://danwang.co/LinkedIn: https://www.linkedin.com/in/danwang15/Twitter/X: https://x.com/danwwangProduction by supermix.io and EPIIPLUS1 Ltd, including Chantal Smith, Marija Gavrilov, Nathan Warren and Hannah Petrovic. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Is AI a bubble? In this episode, I unpack a new five-gauge framework for understanding the biggest question in tech. Drawing on lessons from past manias – railways, telecoms, the dot-com boom – and grounding our analysis in fresh data, we examine economic strain, revenue growth, valuations, and the quality of capital fueling AI’s ascent. This is our effort to cut through hype and fatalism to provide a clear dashboard: where today’s AI build-out looks like a genuine boom, and where early warning signs of bubble dynamics may be emerging. Whether you’re an investor, policymaker, or executive, this framework offers a disciplined way to navigate the noise.Jump to the best parts:(00:29) Echoes of the past (01:31) The 5 gauge framework (01:54) Gauge #1: Investment intensity(03:45) Gauge #2: Monetization level (04:48) Gauge #3: Revenue trajectory (06:23) Gauge #4: Valuation level(07:24) Gauge #5: Quality of capital (10:10) Overall assessmentProduced by EPIIPLUS1 and Supermix. Thanks to my team: Nathan Warren, Hannah Petrovic, Chantal Smith & Marija Gavrilov Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

GPT-5 was the most advanced AI when it was released, but most people were disappointed. Why? In this episode, I unpack the two key paradoxes that shape how we judge new technology: shifting goalposts and negative space.Timestamps: (0:00) The reaction to GPT-5 (0:40) First paradox (2:55) Second paradox (5:29) Why this matters Dig deeper: https://www.exponentialview.co/p/the-paradox-of-gpt-5  Production by supermix.io and EPIIPLUS1 Ltd, including Chantal Smith, Marija Gavrilov, Nathan Warren and Hannah Petrovic. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.