May 25, 2023 · 1h 6m · a16z
The Rise of the Exponential Organization with Peter Diamandis and Salim Ismail
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
Host Steph Smith sits down with Peter Diamandis and Salim Ismail to discuss Exponential Organizations 2.0, exploring how modern enterprises can leverage accelerating technologies like AI, lean operational structures, and a Massive Transformative Purpose to achieve 10x performance over traditional linear incumbents.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The host holds 19% of the talking time here. How this is scored →
The host as informed peer 3.1 Guest teaching 4.8 Guest disagreement 0.4 The host pushing back 0.6
speaking balance: gold is the host, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 49:09 Salim forcefully rejects job loss narrative
Salim explicitly expresses frustration ('I go crazy when I hear this') and rejects the premise that technological automation inherently destroys overall employment.
Hardest push from the host ▶ 44:44 Steph challenges human displacement in ExO modelSteph directly challenges the guests on the human cost of replacing W-2 staff with software algorithms, asking how leaders account for loyal long-term employees.
Biggest teaching moment ▶ 40:48 Salim presents 40x Fortune 100 ExO index performance gapSalim educates the host and audience with empirical 7-year trailing performance data proving ExO-aligned Fortune 100 companies outperformed linear peers by 40x in shareholder returns.
The host holds their own ▶ 55:12 Steph demonstrates practical AI prompting techniquesSteph displays strong individual expertise and practical understanding of generative AI by detailing a multi-prompt experiment exploring paradigm shifts and bias in ChatGPT outputs.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The host as informed peer | Guest teaching | Guest disagreement | The host pushing back | Why |
|---|---|---|---|---|---|---|
| Understanding Logarithmic Scales and Technological Exponential Growth | 0 | 0 | 0 | 0 | Steph Smith introduces the podcast episode in a monologue framing exponential technological growth and introducing authors Peter Diamandis and Salim Ismail. As a monologue intro, host expertise and pushback scores are set to zero. | |
| Official Andreessen Horowitz Legal Disclaimer and Podcast Title Sequence | 2 | 5 | 0 | 0 | Steph opens the interview asking for the definition of an exponential organization. Salim and Peter explain the 10x performance benchmark and cite examples like TED and Singularity University. | |
| What Is Fundamentally New in the ExO 2.0 Era | 2 | 5 | 1 | 0 | Steph asks what is fundamentally new in 2023 compared to a decade ago. Peter and Salim explain how generative AI, 5G, and democratization of compute enable tiny teams to scale rapidly. | |
| Differentiating Linear versus Exponential Organizational Structures | 3 | 6 | 0 | 1 | Steph asks how to differentiate exponential companies from traditional linear organizations in gray areas. Salim explains the economic distinction of dropping supply costs exponentially relative to demand costs. | |
| The Incumbent Dilemma and the Asteroid Extinction Analogy | 5 | 4 | 1 | 1 | Peter introduces his asteroid and dinosaur extinction analogy for incumbent corporate disruption. Steph demonstrates host expertise by bringing up an infographic showing how top 20 companies from 1989 were entirely replaced. | |
| Three Strategies for Incumbent Survival: Edge Units, M&A, and Platforms | 3 | 6 | 1 | 0 | Salim outlines three incumbent strategies for survival (edge units, M&A, platforms) and declares Coase's Law obsolete due to modern web services. Peter emphasizes the necessity of having a crazy idea department. | |
| Unpacking the ExO Framework: The Massive Transformative Purpose (MTP) | 2 | 6 | 0 | 0 | Steph transitions the conversation to actionable steps for building an ExO. Salim and Peter explain the foundational concept of a Massive Transformative Purpose (MTP). | |
| Formulating an Effective MTP and the Purpose-Driven Revenue Premium | 4 | 5 | 0 | 1 | Steph pushes the guests to explain how to distinguish a genuine MTP from vague corporate slogans. Salim gives practical heuristics and shares how Unilever's purpose-driven brands outperformed others. | |
| The SCALE Framework: Unpacking ExO Externalities | 3 | 6 | 0 | 0 | Salim details the SCALE framework for ExO external mechanisms. Steph adds a helpful point about Wikipedia to illustrate community and crowd dynamics. | |
| The IDEAS Framework: Internal Mechanisms for ExO Scale | 5 | 4 | 0 | 0 | Salim introduces the IDEAS acronym starting with interfaces and dashboards. Steph displays strong domain context by contrasting modern software APIs with older human dispatcher models. | |
| Driving Innovation through Experimentation and Autonomy | 3 | 5 | 0 | 0 | Peter details experimentation culture over reliance on expert opinions, referencing Jeff Holden and Astro Teller. Salim discusses Amazon's quarterly experiment tracking metrics. | |
| Radical Organizational Decentralization: The Haier and Tangerine Bank Cases | 4 | 6 | 0 | 1 | Steph raises a timely question regarding flattening organizations and downsizing during downturns. Salim responds with case studies on Haier and Tangerine Bank operating without traditional corporate hierarchies. | |
| Social Technologies in Exponential Organizations | 3 | 4 | 0 | 0 | Salim covers social technologies as the final ExO attribute. Peter and Salim outline their updated book release and their interactive AI chatbot trained on the ExO knowledge base. | |
| Empirical Proof: The Fortune 100 ExO Index Study | 2 | 7 | 0 | 0 | Salim presents empirical findings from a 7-year study of Fortune 100 companies showing a 40x stock return gap. Peter outlines the 6 Ds framework and recounts Kodak's downfall. | |
| Overcoming Organizational Resistance and Innovating at the Edge | 5 | 5 | 1 | 6 | Steph offers direct pushback on the human cost of transitioning to automated ExO models, asking how leaders should handle concerns about existing employees. Salim explains how companies manage innovation at the edge. | |
| Technological Automation, Workforce Expansion, and Abundance | 3 | 6 | 4 | 1 | Peter introduces upcoming humanoid robotics trends. Salim strongly counters common economic job-loss fears, citing the introduction of bank ATMs as a historical precedent for capacity expansion. | |
| Implementing AI Tools and Appointing a Chief AI Officer | 3 | 5 | 0 | 0 | Steph asks how generative AI tools are practically deployed inside modern ExOs. Salim outlines MTP experiment generation, and Peter advises appointing a Chief AI Officer. | |
| ChatGPT Experiment on Societal Paradigms and Mindset Expansion | 6 | 3 | 0 | 0 | Steph demonstrates her hands-on AI expertise by describing a personal ChatGPT experiment comparing altruistic and realistic framings. Peter responds with the concept of a Chief Mindset Officer. | |
| ExO Diagnostic Framework and Community-Driven Business Models | 2 | 5 | 0 | 0 | Steph asks where companies should begin evaluating these attributes. Salim explains their free diagnostic survey and illustrates community leverage with Xiaomi's 1,000 superusers. | |
| Grand Challenges: XPRIZE Innovations and AI-Driven Global Education | 2 | 4 | 0 | 0 | Steph concludes by asking about grand global challenges catching their interest. Peter highlights wildfire detection and longevity XPRIZEs while Salim discusses AI-driven countrywide education in Malawi. |
Statements from this episode (43)
Smith: Instagram had 13 employees when acquired for $1 billion in 2012
“Instagram, for example, had 13 employees when it was sold for a billion dollars, and that was in 2012.”
Ismail: Exponential organizations perform 10x better, faster, and cheaper than peers
“So the definition we have for an EXO or exponential organization, as we call it, is that it's delivering 10 X better, faster, cheaper than its non-EXO peers.”
Diamandis: Digitization enables small teams to build billion-dollar startups in record time
“I mean, the ability to go from an idea to a billion dollar startup in record time with a really small number of individuals, there's something different and it really is the digitization of Every aspect.”
Diamandis: Generative AI is spawning tens of thousands of single-person startups
“Generative AI right now, which is a tool that's spawning not thousands, maybe tens of thousands of new startups you know, individual single person startups out of no place”
Ismail: Reducing marginal cost of supply to zero drives market cap growth
“What we find with exponential organizations, they found a way of dropping the cost of supply exponentially, right? So you think about Airbnb, the cost of adding a room to their inventory is near zero. If you're hired, you have to build a hotel. Same with Uber …”
Diamandis: Legacy companies must build exponential units outside their core
“The challenge is if you're a large established company, trying to retrofit your organization to become an exponential organization is almost impossible. And there's lots of reasons for it, right? You're basically fighting everything you've been doing for the l…”
Diamandis: Heavy legacy organizations will likely not survive the 2020s
“The difficulty is we're going to see such massive disruption in this decade. Right, so if you're a large organization with the layers of management and a massive lead keel of assets and resources that you can't offload during ups and downs you're likely not to…”
Smith: None of 1989's top 20 companies remain in top 20 today
“I saw this infographic that basically showed the largest companies in 1989. The top 10 or top 20, none of them are in the top 20 today.”
Diamandis: Large corporations will survive by acquiring innovation, not building internally
“But it's going to be that the larger corporations are going to be surviving by buying innovation versus creating it inside in-house.”
Ismail: Large companies trigger internal immune responses against self-disruption
“When a big company tries to do anything disruptive, they get an immune system response from the rest of the company and they get attacked.”
Ismail: Tech winners like Apple and Google succeeded by becoming platforms
“Yahoo, BlackBerry, Nokia all failed to become platforms, right? Google, Apple, Facebook, Amazon all successfully became platforms, and that business model basically carries you through because now you wire yourself into the infrastructure.”
Ismail: Coase's Law is obsolete due to cheap web service transaction costs
“So in the book, we actually declare Coase's Law, which has been, ah, very touted for a hundred years, essentially gone. Because the pay, the need to control a big company makes it too slow to navigate all these massive changes that are happening. And B, the tr…”
Ismail: Study of 200 unicorns found 100% had a Massive Transformative Purpose
“The way we came up with this was in 2012 to 2014, we analyzed the 200 fastest growing unicorns in the world and said, how are they doing this? How's Uber scaling so fast or Airbnb or Ted or others? And we found that all of these organizations had what we call …”
Diamandis: Corporate missions focused solely on shareholder return fail talent today
“And unfortunately we see historically a lot of the traditional blue chip companies are like, you know, to provide outsized shareholder returns. I'm sorry, that's just not going to cut it today. A lot of the smartest you know, gen, whatever you want are looking…”
Ismail: Unilever's five most profitable brands were those adopting transformative purposes
“When we wrote the first book about around 20 15, Paul Pullman, who was CEO of Unilever, asked me to come and talk to his top 200 executives. And after that session, he ordered every brand in Unilever to take on an MTP. And every brand had to be purpose-driven …”
Diamandis: Rigid Salaried Workforces Risk Bankruptcy From External Shocks
“If you have a huge workforce that is on your salary base and you don't have any agility to flex up and flex down, You can be bankrupted by an externality like a pandemic or like whatever it might be.”
Ismail: The launch of AWS marked the birth of exponential organizations
“Cloud computing was probably the most relevant here in the kind of we considered the birth of an EXO when Amazon Web Services launched, because now you can take computing off the balance sheet and make it a truly variable cost.”
Ismail: NFT Projects Build Multi-Billion Dollar Ecosystems as Exponential Organizations
“And now with crypto economics and token economics all of these NFT projects are basically EXOs. They have a big purpose. They create crypto economics and they reach multi-billion dollar ecosystems incredibly fast. Just by incentivizing them in the right ways.”
Ismail: Only 3.5% of large global companies use OKRs
“So we worked out that only three and a half percent of big companies around the world use OKRs.”
Ismail: Five-year strategic corporate plans are obsolete by completion
“Because for big companies, by the time they finish their five-year plan, it's out of date.”
Ismail: Amazon penalizes employee bonuses if their experiments do not fail enough
“We saw, for example, an Amazon that every team is measured on how many experiments they ran this quarter, how many succeeded, how many failed. If you're not running enough experiments and not enough for failing, you're actually, your bonus gets dinged, right?”
Ismail: Valve Software operates with zero management hierarchy or meetings
“Like valve software in Seattle that has no CEO or reporting lines, job descriptions, management meetings of any kind.”
Ismail: Haier split 80,000 employees into 2,000 autonomous micro-enterprises
“They make fifty-five million fridges and ovens a year. There were 80,000 people in a classic pyramid structure, command and control, top-down, etc. CEO one day wakes up and goes, can't meet my corporate goals using this structure. Blows it up. Turns 80,000 peo…”
Ismail: GE sold its appliance business to Haier after failing to compete
“GE actually gave up and sold GE appliances to them because they couldn't compete.”
Ismail: Tangerine Bank operates without traditional reporting lines or job descriptions
“There's a Canadian bank called Tangerine Bank, which used to be ING Direct. Which actually operates like this. So in a regulated bank, there's no CEO reporting lines, job descriptions, et cetera.”
Salim Ismail built an AI chatbot to interface with Exponential Organizations II
“So over the last three months, we've been building an AI chatbot to interface with the book.”
Salim Ismail: Books will be continuously updated like software
“And we think the future of books is going to be like this. A book that was, took a few years to write, et cetera, will be out of date if it's not being constantly updated and refreshed.”
Ismail: Top exponential Fortune 100 companies outperformed bottom 10 by 40x
“Over a seven-year period, the top 10 did better than the bottom 10 by three x on revenue growth 6.4 times on profitability, 10.9 times on return on assets, but shareholder returns, compound annual growth rate, top 10 outperformed the bottom 10 by 40 x.”
Ismail: Top 10 ExO companies responded to COVID-19 pandemic five times better
“And what was really incredible was, ah, when the pandemic hit, these, the top 10 did five times better than the bottom 10 in responding to the pandemic, right?”
Diamandis: In 1996, Kodak had a $28B market cap and 140,000 employees
“So Kodak in 1996, the top of its game, the twenty eight billion dollar market cap, 140,000 employees.”
Ismail: Companies without generative AI will not survive
“If you're not building generative AI into your products and services going forward, you're not going to last.”
Ismail: Apple's real competitive edge is its organizational design, not products
“I will argue that Apple's real real edge is organizational, because what they will do is they will form a small team that's really disruptive. They will put that team at the edge of the company. They'll keep them secret and stealth, and they'll say to them, go…”
Ismail: Google missed early LLM deployment due to risk aversion
“Google was basically the foundation of LLMs, right? They were too nervous and careful and risk averse to launching.”
Ismail: ATMs lowered branch costs 10x without reducing bank teller jobs
“When we looked in the past, when we've had disruptive technological innovation,
To come in and completely clean up.
We found that we don't lose jobs.
What we do is we increase capacity.
I'll give you a very clear example, which is in the seventies, we created …”
Ismail: Countries with highest robotics adoption maintain the lowest unemployment rates
“The most roboticized companies in the countries in the world that have the highest penetration of robotics are South Korea, Sweden, Germany, which have the lowest unemployment.”
Diamandis: Proliferation of AI and robotics will lead to Universal Basic Income
“And so we probably will have some version of universal basic income. Coming down as more robots and AIs come online”
Ismail: Generative AI will replace most lawyers
“The best benefit is we'll replace most of the lawyers. Because you could just say, generate me a contract that has these characteristics and it'll throw in the clauses and et cetera.”
Diamandis: Every company must hire a Chief AI Officer
“Every company needs to have what I call a chief AI officer. It is someone who understands what's going on in the AI universe and is sitting next to the CEO as a strategic advisor to say, We need to be injecting it here, here, and here, and partner with this or…”
Ismail: Smartphone maker Xiaomi incurs zero marketing or promotional costs
“Then you have Xiaomi in China, the second biggest phone company, and they're successful because they have a thousand super users that vote on what features should go into the new phone, and they get the phone early and can boast to people, look, I voted, and a…”
XPRIZE launches competition to detect and extinguish wildfires within 10 minutes
“We just launched an XPRIZE for wildfires. You can monitor a thousand square kilometers, detect a wildfire, add ignition, and put it out 10 minutes automatically.”