Jul 9, 2026 · 43m · american-optimist
Why AI is Bigger Than the Industrial Revolution
Joe Lonsdale · 12m spoken JoeBen Bevirt · 9m spoken JD Ross · 5m spoken Scott Wu · 4m spoken Alex Kolicich · 4m spoken Russell Kaplan · 1m spoken
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
Hosted by venture capitalist Joe Lonsdale, this episode of American Optimist investigates how artificial intelligence is driving a new industrial revolution through candid conversations with software founders, aerospace pioneers, and investment partners. The discussions explore autonomous coding agents, electric vertical aviation, institutional adoption hurdles, and historical macroeconomic parallels.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Joe holds 30.7% of the talking time here. How this is scored →
Joe as informed peer 5.3 Guest teaching 3.5 Guest disagreement 1.3 Joe pushing back 2.3
speaking balance: gold is Joe, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 13:06 JD Ross dismisses assumptions about foundation model boundaries
JD flatly rejects conventional thinking that foundational AI labs will avoid vertical enterprise SaaS, insisting they will aggressively capture the highest-value application surfaces.
Hardest push from Joe ▶ 39:50 Joe Lonsdale challenges the managed displacement thesisJoe refuses Alex Kolicich's premise that technological job churn should be managed, arguing that political populists inevitably abuse transition mandates to seize power.
Biggest teaching moment ▶ 2:44 Scott Wu corrects host misconception on agent evaluation metricsWhen Joe wonders why models take human hours rather than running instantly, Scott explains that benchmark hours represent the human labor equivalent accomplished between interventions.
Joe holds their own ▶ 11:44 Joe Lonsdale deconstructs 100x software productivity claimsJoe pulls from his direct experience spending nearly $2 billion engineering Addepar to challenge simplistic assertions that AI will instantly drop complex software build costs by 100x.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Joe as informed peer | Guest teaching | Guest disagreement | Joe pushing back | Why |
|---|---|---|---|---|---|---|
| Cognition: The Exponential Scaling of Autonomous AI Agents | 4 | 5 | 1 | 2 | Joe asks about the horizon for autonomous coding agents but reveals confusion regarding METR benchmark metrics. Scott Wu politely clarifies that benchmark hours denote human-equivalent labor time between interruptions, not model execution run-time. | |
| Cognition: AI as an Engine for Small Business Growth | 5 | 2 | 0 | 0 | The exchange is highly cooperative as Russell and Scott explain Cognition's philosophy of hiring former founders. Joe reinforces this by referencing his early recruitment days at Palantir and his prior history hiring Scott. | |
| Coverage: The End of Manual Coding and the Agentic Enterprise | 5 | 3 | 1 | 1 | JD Ross describes how Coverage engineers use autonomous agents to turn around customer software requests within 24 hours. Joe readily engages with product prioritization dynamics based on his own entrepreneurial background. | |
| Coverage: The SaaS Barbell Effect and Defensible Domain Moats | 7 | 5 | 3 | 5 | Joe challenges inflated claims of 100x productivity gains by citing actual engineering spend and development economics at Addepar. JD counters Joe's assumption that foundational model labs will avoid enterprise SaaS, arguing labs will capture top-value workflows in a barbell distribution. | |
| Coverage: Enterprise Adoption Bottlenecks and Frontier Model Integration | 5 | 4 | 1 | 2 | Joe questions whether foundation model labs have the chops to handle enterprise friction. JD explains that enterprise absorption capability and context delivery are the genuine bottlenecks, laying out their evaluation threshold rule. | |
| Joby Aviation: Autonomous Airspace, NVIDIA Hardware, and Pilot Training | 4 | 3 | 1 | 3 | Joe probes the commercial contradiction between developing fully autonomous aircraft and aggressively recruiting human pilots. JoeBen explains the regulatory reality and Joby Academy flight simulator training developed with CAE. | |
| Joby Aviation: Point-to-Point Urban Transit and Rooftop Infrastructure | 5 | 5 | 1 | 2 | Joe jokingly raises concerns about explosive hydrogen like the Hindenburg disaster. JoeBen explains the technical mechanics of cryogenic liquid hydrogen fuel cells and the 6x net efficiency advantage over conventional jet fuel. | |
| Joby Aviation: Unleashing Superhuman Aerodynamic Engineering with AI | 4 | 2 | 0 | 0 | A collaborative segment where JoeBen recounts how AI tools granted Joby's chief aerodynamicist superhuman productivity. Joe enthusiastically affirms the thesis that American engineering cycles are dramatically compressing. | |
| 8VC: Dissecting the AI Bubble and Institutional Diffusion | 6 | 3 | 2 | 3 | Alex Kolicich and Jack dismiss talk of an AI valuation bubble by pointing to reasonable forward multiples like Nvidia at 25x. Joe pushes back slightly, arguing that once widespread productivity improvements become undeniable, speculative bubble dynamics will follow. | |
| 8VC: Macroeconomic Realities and Historical Infrastructure CapEx | 6 | 4 | 1 | 1 | Jack highlights Micron's flat valuation multiple despite a 250% stock run, while Alex cites historical CapEx data from Russian and Chinese industrial revolutions. Joe draws parallels to late 19th-century American infrastructure waves. | |
| 8VC: Managing Workforce Disruption and Reforming Retraining | 7 | 3 | 3 | 6 | When Alex suggests that societal transitions must be actively managed to protect disrupted workers, Joe pushes back hard against bureaucratic intervention, warning that populist leaders exploit such mandates. Joe also cites $40 billion spent on ineffective federal job retraining programs. |
Statements from this episode (32)
Scott Wu: Agent-forward engineers see 5x to 10x productivity gains
“I think right now, you know, you have this core group of, we'll call it like agent forward engineers, right? Or agent forward companies that understand how to use this and they are seeing these, you know, five X, 10 X productivity games as a result.”
Scott Wu: AI autonomous work time doubles four to five times yearly
“And the thing that's really crazy about this stat is, you just see it very consistently double. And I think for the last few years, it's doubled about four or five times every year, you know, which is insane, which means, you know, you wait two or three months…”
Scott Wu: AI coding will shift toward proactive, event-driven autonomous work
“So you're gonna see a lot more, I think, proactive work. You're gonna see a lot more basically like event-driven kickoff work where it's, it doesn't have to be human that's moderating it every step of the way.”
Kaplan: AI will trigger an explosion in new small businesses
“I actually think one thing I'll go on the record on is I think there's gonna be an explosion in small businesses.”
Kaplan: Every engineer on Cognition's special projects team is a former founder
“We have one team at the company they're called special projects engineers. And basically every person on that team is a former founder, like every single one.”
Scott Wu: Testing candidates on tasks AI can do is the wrong interview strategy
“And I think if you're asking the question of like, how do we evaluate people on exactly the thing that AI can already do, that's kind of the wrong way.”
Ross: Zero full-time engineers at Coverage write code manually anymore
“Not a single one of our full-time engineers right now writes code anymore. They do everything through Codex, Cloud Code, or Cursor.”
Ross: Coverage uses automated AI agents to one-shot PRs from Linear tickets
“Cole, who runs our engineering team, has like six agents that are running at once at all times. Just doing different kinds of projects. When we create a new ticket in linear, it automatically kicks off this thing that basically creates a dev environment, which…”
Ross: The cost of building new software is asymptotically dropping to zero
“I think right now, like the cost of software is just asymptotically dropping to zero, a building new software, which means that the way you build a company completely changes.”
Ross: Coverage built and shipped a custom enterprise feature in one day
“Like we can now have an enterprise customer come up who says, oh yeah, this is so close, but we actually need it for is X. We can build X in one day, ship it live and close that deal. We did this last week.”
Ross: Software development has fundamentally shifted from writing to editing
“If the job like in the past was writing, now it's editing and shifting from being a writer to an editor is a very I think for some people they're thriving in it and some people are really shocked by it.”
Ross: Foundation models will build apps targeting high-volume token usage
“Anthropic, OpenAI, Google, they're all gonna, in the same way that you probably don't want to build an email company while Google's on its arise, because they're gonna come and eat that, like, every one of these companies is looking at, where are our tokens be…”
Ross: In five years, workers will one-shot custom software, making SaaS worthless
“If you take the five year view, kids who are in college today who come out and start working in a construction company or something are going to look at every problem they're paying for software and say, this doesn't do exactly what I need. And I can one shot …”
Ross: Model companies are determined to capture software development tooling
“The first one obviously is anything that's adjacent to software development. They're like, they've decided that they're going to war to win that. And I think that's a really hard place to, you have to be.”
JD Ross: Foundation model labs will attack vertical professional services software
“And I think over time I would be surprised if more of that doesn't get attacked by The large AI foundational companies.”
JD Ross: Enterprise absorption will bottleneck AI adoption for years
“The ability to absorb and adopt the tools is going to be the limiting factor for years to come. And I think that the capabilities of these models is already so far ahead of our ability in most enterprises to just absorb and adopt change that that is like the w…”
JD Ross: All WithCoverage emails are drafted by AI before human review
“All of our, all the emails that are, that we send are drafted by AI before we send them and a person reviews them and sends them out.”
Bevirt: eVTOL aircraft can compress two-hour commutes to ten minutes
“It's you know, being able to take a trip that, that, you know, takes you half an hour on the ground and turn into a five or 10 minute trip, you know, and sometimes, sometimes you're replacing a two hour trip because you've got traffic.”
Bevirt: Joby flew a hydrogen-electric VTOL aircraft over 500 miles
“We flew a hydrogen electric version of our VTOL aircraft more than 500 miles”
Bevirt: Hydrogen fuel cells yield a 6x efficiency gain over jet fuel
“Three times lighter the hydrogen is three times lighter than jet fuel. And then we can convert it into propulsion twice as efficiently. So you get a six X gain.”
Bevirt: Hydrogen will redefine every class of commercial and military aircraft
“I think that over time hydrogen propulsion will redefine every class of aircraft that's flown today, both on you know, on the commercial side and for defense applications.”
Bevirt: Over 90% of Joby aircraft components are made in-house
“More than 90% of the components on the aircraft. Are designed and manufactured in house.”
Bevirt: Joby's chief aerodynamicist is 10x faster using AI
“He's now able to build in an afternoon
What would have previously taken him weeks or months?
he's like, I'm 10 X per 10 X as productive as I was just a few months ago.”
Bevirt: Joby's research team designed aircraft 10x better than current planes
“These are the folks that have a book of a dozen new aircraft designs that not just are like, 10% better than aircraft that we have today, but, you know, five and 10 times better.”
Kolicich: AI is not a bubble and is probably still under-hyped
“And I think it's an emphatic, no, it's not a bubble. Probably it's still a little under hyped.”
Kolicich: Existing AI models are sufficient to trigger an economic revolution
“Even if we do that, you'll still have a revolution in the economy. You'll still have a revolution in the way things happen because the, Limiting factors now are not the technology, it's diffusion of the technology into the economy.”
Kolicich: AI will force the legal industry to abandon billable hours
“That's gonna be, law's gonna be fully transformed. From really the bottom up. Like, their whole business model will have to shift. So you'll have to move from today law is sold per hour. You're gonna have, hours don't make any sense when it's agents doing work…”
Kolicich: AI investment at 4% of GDP lags historical industrial revolutions
“If you look at the Industrial Revolution in Russia in the early 20th century, I think investment as a percentage of GDP got up to 20 or 25%, and then China was investing 40 or 50% in their industrial revolution of GDP, and now in AI we're investing, like, thre…”
Lonsdale: Populists will mismanage the AI workforce transition
“What's gonna happen is some crazy populists, both on the right and left, this is like a non-political comment, crazy populists want to manage it because they want the power, and then they're just gonna screw it up, right? And they're gonna steal.”
Lonsdale: The US spends $40B annually on incompetent job training programs
“We have forty billion dollars a year of training programs, and they're just completely incompetent.”
Lonsdale: Apps and services layer companies are safer than infrastructure
“This tends to happen in every cycle where the infrastructure gets overdone at some point. So that, but I think this, well, this is again, speaking to our book is I think some of the apps and services layer companies are a lot safer when that happens”