Bold

By Peter H. Diamandis & Steven Kotler

9 min read

Why does it matter? Because the instinct to think small is the most dangerous risk you'll take.

Here's a counterintuitive thing about ambition: the cautious version is actually the more dangerous bet. Shoot for 10% improvement and you're in a crowded race against everyone with your same skills, your same tools, your same assumptions. Shoot for ten times better and almost nobody's competing with you — because most people assume it's impossible. Bold is the book that dismantles that assumption piece by piece. Exponential technologies have quietly handed ordinary entrepreneurs the leverage that once belonged only to governments and billion-dollar corporations — but the technology is only half the equation. The other half is psychological: specific ways of setting goals, building teams, and harnessing crowds that make moonshot-scale problems not just approachable but, by some measures, easier than the incremental alternative. Miss that second half and the leverage stays theoretical.

The Dinosaurs Didn't See It Coming — And Neither Will You

In 1975, a young Kodak engineer named Steven Sasson built something that should have made his company immortal. The device weighed eight and a half pounds, looked roughly like a toaster, and captured images at 0.01 megapixels — so blurry the pictures were nearly useless. It stored thirty black-and-white images on a cassette tape and took twenty-three seconds to snap a single shot. When Sasson demonstrated it for Kodak executives, they didn't ask how it worked. They asked when it would be a problem. Fifteen to twenty years, Sasson told them, running the numbers on Moore's Law. The executives thanked him and quietly shelved it to protect the company's chemical and photographic paper margins. In January 2012, Kodak filed for bankruptcy. Three months later, Instagram — thirteen employees, zero film, zero paper — sold to Facebook for a billion dollars.

Kodak's failure wasn't stupidity. It was something more dangerous: an inability to see exponential change as change at all. There's a name for the mechanism — the Six Ds, a predictable chain reaction that begins the moment something goes digital. First comes Digitalization, when a physical process converts to ones and zeros and hops onto an exponential growth curve. Then comes Deception, the phase that kills companies: early doublings look like nothing. Sasson's 0.01 megapixels doubles to 0.02, then to 0.04. To any executive watching quarterly results, those numbers round down to zero. But thirty doublings from any starting point produces a billionfold increase. By the time the numbers look serious, the window to respond has already closed. Disruption follows, then Demonetization (the revenue stream vanishes — nobody buys film when pixels are free), then Dematerialization (the standalone camera disappears into the phone in your pocket), and finally Democratization, when the technology becomes cheap enough to reach everyone.

The pattern is the point. Disruption doesn't arrive as a crisis you can see and respond to. It arrives as a long period of apparent nothing, followed by an outcome that looks sudden only because you were measuring the wrong things. Kodak's executives weren't watching megapixels double. They were watching film sales. By the time one chart looked alarming, the other had already won.

The Tools of Planetary-Scale Change Now Cost Less Than a Used Car

Here is the actual state of things: the tools that can reshape industries at planetary scale are already commoditized, sitting in the cloud, available to anyone with a credit card and a clear enough idea of what to do with them. The bottleneck stopped being money some time ago. It is now vision.

The moment that unlocked this shift happened in 1993, when a twenty-two-year-old undergraduate named Marc Andreessen released a browser called Mosaic. The internet itself had been running since the early 1970s — a text-heavy network built for scientists and military researchers who knew Unix commands. Technically, the infrastructure was there. But in early 1993, the entire web consisted of twenty-six sites. What Andreessen did wasn't invent new infrastructure; he built a graphical front door. Swap the command line for point-and-click, add images, make it run on Windows — the operating system on roughly eighty percent of the world's computers — and suddenly the same technology locked inside research institutions was accessible to anyone. By mid-1995, there were more than ten thousand websites. By the end of 1998, several million. The technology hadn't changed. The interface had.

The lesson Diamandis and Kotler draw from this is sharp: the signal that an exponential technology is ready for entrepreneurs isn't raw capability — it's the moment a non-expert can sit down and use it. Watch for the interface, not the lab announcement.

That same door is now opening across half a dozen fields at once. Autodesk CEO Carl Bass describes it through what he calls infinite computing: a problem that once demanded ten thousand seconds on a single processor now takes one second when you throw ten thousand processors at it in parallel, at identical cost, because Amazon and Rackspace rent that computing power by the minute. Pharmaceutical companies already use this to simulate millions of drug-protein interactions overnight — trial-and-error at a scale that would have required decades of wet-lab time. Experimentation, the thing that always cost money and time, now costs almost nothing. Mistakes are cheap. Trying every option is cheaper than choosing carefully among a few.

For a founder in a garage, that means waking up with the same computational leverage that once required a government defense budget. The era of needing to pick your one best shot is over. You can run a hundred experiments simultaneously, let the data collapse the field, and build the thing that survives. That is what cheap enough to try everything actually looks like — and it is genuinely new.

Thinking Small Is Actually the Riskier Bet

Which is actually riskier: shooting for a 10 percent improvement, or shooting for something ten times better? The instinctive answer is obvious. The instinctive answer is wrong.

Astro Teller, the man Google put in charge of its moonshot laboratory, has a mathematical argument that cuts against most founders' intuitions. Aim for 10 percent better and you're doing something precise and constrained: starting from the current state of the art, inheriting all its assumptions, using the same tools and processes as every other competitor, entering what is a global smartness contest. Statistically, you will lose that contest. Someone somewhere has more engineers, more capital, more institutional knowledge of the same incremental problem. A 10 percent target is a crowded target.

Aim for 10x and something counterintuitive happens. The rulebook dissolves. You cannot get ten times better by tweaking existing assumptions; you have to abandon them entirely, which means you're no longer competing on the same terms as everyone else. The field empties out. The best people, drawn to the hardest problems, show up. And when reality erodes your ambitions — as it always does — you still land somewhere remarkable. The modest goal never offered that guarantee.

Elon Musk runs the same logic at the level of individual assumptions. When the battery industry quoted him a price per kilowatt-hour for Tesla's energy storage, he didn't accept the number as a fact about the world. He asked what a battery actually is — its physical constituents, nothing more. Carbon, nickel, aluminum, some polymer separators, a steel casing. He looked up the spot prices for those raw materials on commodity exchanges and ran the arithmetic. The floor came out near $80 per kilowatt-hour, far below the price the industry treated as fixed. The entire consensus had been reasoning from the existing product rather than from first principles, and in doing so had collectively agreed on a cost ceiling that didn't actually exist. One physics question dissolved an industry assumption.

Both moves are the same move at different scales: abandon the inherited map and navigate from the terrain. Modest goals don't just aim lower. They lock you into the existing map, hand the competition a head start, and remove the creative pressure that forces genuinely new thinking. Ambition, it turns out, is not the risky choice. It's the one that clears the field.

Before You Shoot for the Moon, Build Your Launchpad

Think of the medieval folktale about soldiers who want to feed a hungry village. They drop three stones into a pot of boiling water and announce they're making stone soup. Curious villagers gather. Someone offers potatoes. Another adds carrots. A third contributes barley. The soldiers didn't have soup — they had the idea of soup, plus enough theatrical confidence to make everyone else want to contribute. That's how you fund a moonshot when you don't know any billionaires.

When Diamandis and two fellow graduate students decided to build an International Space University in the early 1980s, they had no campus, no endowment, and no faculty. What they had was a prior win: SEDS, a global student space organization they'd already built, which gave a small network of people reasons to trust them. That was stone one. Rather than pitch the university directly, they organized something more modest — a conference to study whether such a university was even feasible. That framing mattered: lower perceived risk means easier yes. Stone two.

Here's the pattern you're about to watch: earn trust at one level, use that trust as currency to unlock the next level up. Each commitment makes the next one feel safer to make.

Then came the leverage move. A contact secured the head of the Canadian Space Agency as a speaker. That single commitment recruited the European Space Agency. Having both agencies on the roster brought in the Japanese, then the Russians, then NASA. Nobody wanted to be the one missing from a table that everyone else had already joined. By the time the conference opened, it looked less like three grad students with a slide deck and more like an international space summit — so credible that the feasibility conference became the founding conference on the spot. Within a weekend they had seed capital for a summer program.

Five summer programs across five countries, roughly five hundred and fifty alumni, a real track record — and suddenly a request for proposals asking cities to bid on hosting a permanent campus drew seven responses worth between twenty and fifty million dollars each. Strasbourg eventually built them a fifty-million-dollar facility.

Audacity and sequencing aren't opposites. The moonshot is the destination; the staged approach is what makes the journey believable to everyone you need to bring along.

The Crowd Is the Superpower You Haven't Hired Yet

In 2012, a small group of freelance filmmakers accepted a challenge brief from Colgate Palmolive: produce a thirty-second deodorant ad for Speed Stick, digital placement only, total prize purse $27,000. They delivered something so unexpectedly sharp that Colgate bumped it into a Super Bowl slot — 110 million viewers, the most expensive advertising real estate on the planet. On USA Today's post-game ad meter, the spot finished 24th out of 60. The ads it beat had budgets 500 times larger, produced by agencies with decades of blue-chip client relationships and entire floors of creative directors. The crowd didn't just show up. It won.

Most people's mental model of crowdsourcing stops at cheap, simple labor — transcribing audio, tagging images, writing product descriptions at scale. That's real, and useful. But it misses what's actually happening. Tongal, the platform that ran the Speed Stick challenge, structures its contests in phases: a brief ideation round where anyone can pitch a concept in a few sentences, then a production round where directors bid on the winning ideas. A typical campaign generates over 400 distinct concepts before a single frame is shot. The traditional agency model might deliver seven to ten pieces of finished content after months of development. The math isn't close.

The crowd isn't a bargain bin. It's everyone who couldn't get inside the building. Tongal's founder spent years at Paramount watching the same small circle of gatekeepers decide what got made. The Da Vinci Code nearly slipped through their fingers because someone decided a global bestseller had no entertainment value. The frustrated talent was always there. The platform just gave it a door.

Give people an open-ended brief with a meaningful purse and a structured process, and they'll produce Super Bowl commercials. The bottleneck was never talent — it was always the architecture of the ask.

The Prize Is the Product: How to Make the World Solve Your Problem

The math behind the 1927 Orteig Prize tells you everything. Raymond Orteig offered $25,000 to the first pilot to fly nonstop between New York and Paris. Nine teams, sponsored by corporations chasing publicity rather than returns, collectively spent $400,000 chasing it. Sixteen dollars of effort for every dollar on the table — and the resulting media frenzy didn't just crown a winner, it convinced a skeptical public that commercial aviation was real. The $300 billion global aviation industry grew from that single contest. Orteig paid nothing to the teams who failed, which is the structural secret: prizes are efficient because losers self-fund and the sponsor pays only for demonstrated results. A well-designed incentive competition isn't a marketing stunt — it's a leverage machine.

The deeper mechanism is who shows up. Incumbents, bloated with existing revenue to protect, mostly stay home. The prize attracts people with nothing to lose and everything to prove — outsiders who haven't yet learned what's impossible. When the 2010 BP oil spill exposed that offshore cleanup technology hadn't improved in the twenty years since the Exxon Valdez disaster, Wendy Schmidt funded a competition asking teams to beat the industry's best oil recovery rate. A team that had never worked in oil — led by a Las Vegas tattoo artist who tested his prototype in a backyard Jacuzzi — showed up at the official testing tank and more than doubled the industry record. BP's own engineers, with their institutional knowledge and nine-figure budgets, had never cracked that ceiling. The tattoo artist did it in months, with a fraction of the resources, because he had no preexisting assumptions to protect.

One design principle determines whether a prize produces a genuine breakthrough or a clever workaround: define the finish line, never the path. Richard Feynman once offered $1,000 for a working motor small enough to fit inside a one-millimeter cube, intending to accelerate research into what would become nanotechnology. A graduate student built the motor using jeweler's tweezers and a microscope — technically legal, completely useless to the field. Feynman paid up and got nothing he wanted. The rules are the DNA. State the result you need, measure it objectively, and let solvers surprise you with methods you'd never have specified.

The Question the Book Leaves You With

Every tool in this book — the diagnostic frameworks, the staged launches, the crowd platforms, the prize structures — is just architecture for redirecting effort that already exists but has no door to walk through. The tattoo artist wasn't a fluke. He was what happens when a well-designed finish line replaces a credentialed gatekeeping system. Which means the real question isn't whether you have access to exponential tools — you do, today, cheaper than a used car — it's whether you're willing to state your problem in public, define success in measurable terms, and let the solution come from somewhere you didn't expect. The barrier you've been respecting most is probably the one most worth testing. Not because audacity is a virtue, but because the math keeps showing that the people with the least to protect are the ones who actually move things forward on the hardest problems.