Podcast Summary: "My Outlook for 2026 – Orchestration, the Human Edge, and the AI Bubble"
Podcast: Azeem Azhar's Exponential View
Host: Azeem Azhar
Episode Date: January 16, 2026
Overview
In this solo episode, Azeem Azhar presents his forecast for 2026, reflecting on what he calls a fundamental shift in how artificial intelligence (AI) amplifies productivity and alters the nature of work, business, and meaning itself. He introduces three "anchors" shaping the future: the transformed act of making (or building), the pursuit of meaning and authenticity in a world of abundant, cheap creation, and the foundational considerations of energy and capital supporting this exponential change. The episode is rich with personal anecdotes, practical examples, and industry context, offering both optimism and caution for the year ahead.
Key Discussion Points
1. From the "To-Do List" to the "Done List" – The Building Revolution (00:00–22:53)
- AI Maturation: AI tools have crossed the “uncanny valley” (00:40), now enabling individuals to achieve the output of large teams, not metaphorically, but “in terms of actual velocity” (Azeem Azhar).
- Practical Example: Azeem describes building three custom Mac apps to organize 4,000 messy music files, enhance metadata, and generate playlists—accomplished in about 90 minutes without professional developers (04:05–06:45).
- Growth of Custom Tools: He notes that he has created 30–40 such custom apps recently, a trend mirrored in industry:
- Boston Consulting Group’s “consultant as creator” model produced 36,000+ custom tools using AI (17:34).
- The lead developer for Anthropic’s Claude Code reported 100% of December 2025’s code authored by AI (18:10).
- Changing Role of Software Engineers: The translation of needs to code is becoming obsolete; the value is in formulating the right problems and intents.
- “The bottleneck shifts from writing the code to: do you have a good enough backlog that you can articulate, and can you formulate that problem and evaluate it?” (Azeem Azhar, 15:24)
- Rise of "Vibe Coding": Inspired by Andrej Karpathy, "vibe coding" lets users conversationally direct AIs to build tools “vibing” their way through tasks (12:39).
- Shift for Larger Organizations: Large organizations struggle due to linear, approval-based processes—remaining in the “to-do list era,” while nimble teams embrace outcome-based, immediate results (13:40).
- Agentic Coding: The movement from single-agent to multi-agent (parallel) systems means orchestrating swarms of AI agents for specific needs.
2. The Emergence of Orchestration and the Human Edge (22:54–37:18)
- Orchestration Over Creation: The new skill is “knowing which agent does what, how to direct them, and how to evaluate the outcomes”—akin to conducting an orchestra rather than playing every instrument (21:10).
- AI Models’ Self-Awareness: Azeem recounts a Claude agent telling him to use ChatGPT for a task it couldn’t handle, affirming a future of specialized, collaborative AI models (23:28).
- “Don’t use me for this, go and use ChatGPT because it’s tougher and more cynical than I am.” (Azeem Azhar, quoting Claude, 23:40)
- Usage Gap: Data shows even “top” ChatGPT users only had 283 chats in 2025, indicating that true AI orchestration is still rare (26:44). The gap between dabblers (“fondue set owners”) and power-users (“conductors”) is widening.
- Human Value in AI Workflows:
- “It’s about architecture, intent… verifying: have I got where I wanted to get to? What have I learned… and what expertise do I now need?” (Azeem Azhar, 29:44)
- Authenticity as Scarcity: As AI makes synthetic “authenticity” cheap, real authenticity—meaning, experience, and the “human edge”—will become scarce and valuable:
- Synthetic output is “mid copy,” lacking the “interiority” and detail of lived experience (32:55).
- “In a world where everything becomes perfect, we will start to value the complement of perfection, which is the imperfect, the thing that has some meaning.” (Azeem Azhar, 32:02)
- Proof of Work/Authenticity Parallel: “Difficulty plus authenticity will turn into scarcity and value” (35:27), much as Bitcoin’s value derives from proof of work.
- Rise of Community and Taste-Makers: Those who curate, share perspective, and create human connections will find new opportunity.
3. The Foundations: Energy, Economics, and the AI Bubble (37:19–47:15)
- Energy as the New Bottleneck: The entire AI boom is predicated on accessible, affordable energy—especially electricity.
- The US is repurposing gas turbines for data centers as a short-term fix, but solar energy costs are dropping fast due to scaling (“learning curve” effect) (38:17–40:16).
- Doubling of solar deployment predicted for 2026, with global south countries outpacing previous expectations (41:18).
- “Electricity is turning into a manufactured technology because of the learning curve that attaches to solar.” (Azeem Azhar, 38:41)
- Geopolitical Shift: With major oil producers in conflict but oil at just $56 per barrel, strategic leverage is shifting to the “electron” (44:30).
- AI Investment Cycle – Is It a Bubble?: Despite bearish sentiment in late 2025, fundamentals still signal investment boom rather than speculative bubble.
- Memory (HBM) for AI is sold out for 2026, GPU spot prices are rising, and demand is surging among frontier companies (45:05–46:10).
- “This is not over supply. Once these systems get in front of tens of thousands of employees, that’s an enormous step change increase in demand.” (Azeem Azhar, 46:10)
- Outlook: No “bubble top” is anticipated before Q3 or Q4 2026; continued vigilance is warranted.
Notable Quotes & Memorable Moments
-
On the "Done List" Era:
"We're out of the to-do list era. We're into the done list and that's the capability things like Claude Code are providing to all of us."
(Azeem Azhar citing Tom Tunguz, 03:47) -
On Being an "Orchestrator":
"You do need to understand those systems. The agents are starting to understand their own limitations… we're not going to have that singleton AI… it's that society of AI."
(Azeem Azhar, 23:10) -
On the Usage Gap:
"Owning a fondue set but using it every few years is not the same as regularly orchestrating AIs in your workflow… the gap between fondue set owning AI users and users who've figured out orchestration is now a chasm."
(Azeem Azhar, 27:30) -
On Human Work and Authenticity:
"Proof of work that makes authenticity valuable… the hard human work, the work with your pen… the effort that you put in becomes a differentiator."
(Azeem Azhar, 34:50) -
On Solar Energy Economics:
"Electricity is turning into a manufactured technology because of solar’s learning curve…That will fundamentally transform how we produce our electricity. 2026 is going to be another important year for that."
(Azeem Azhar, 38:41)
Timestamps for Important Segments
- 00:00–03:45 – Introduction to 2026, personal AI workflow epiphany
- 04:05–06:45 – Detailed example: building personalized DJ tools with AI
- 12:39–15:24 – Vibe coding, agentic productivity, value shifts from coding to question-formulation
- 21:10–23:40 – Orchestration, multi-agent systems, “society of AI” concept, AI self-awareness
- 26:44–28:50 – Analysis: diffusion gap between experimenters and orchestrators
- 32:02–35:27 – Meaning, authenticity, and human edge in an AI world
- 38:17–41:18 – Role of solar energy, learning rates, global deployment
- 45:05–47:15 – AI bubble analysis; investment, memory/GPU demand, and future risk signals
Final Thoughts
Azeem Azhar paints a picture of 2026 as a pivot point: the “done list” era, where AI acts as a force multiplier for those capable of orchestrating its potential, widening the gap between dabblers and power users. The value for humans has moved upstream—to authentic insight, creative intent, and orchestrating a society of AIs—while foundational elements like energy and capital dictate the sustainability of this new ecosystem. Although talk of an “AI bubble” lingers, evidence still points to a sustained investment boom—at least for now.
For a deeper dive, track Azeem’s ongoing forecasts at "Boom or Bubble AI" and follow Exponential View for continued analysis.