Why Good Ideas Are Rare

Why Good Ideas Are Rare

Understanding Scarcity in the Age of AI

Time to Complete: 15 minutes

PDF 5-Minute Warm-Up Activity can be downloaded above.

Who This Is For: This lesson is for anyone who has to decide whether an idea is worth pursuing, and when. That includes first-year college students encountering innovation economics for the first time, but also the entrepreneur weighing whether to launch now or hold off, the venture capitalist trying to understand why a deal has no real competition, the product manager wondering why the obvious solution has not been built yet, the R&D director deciding how long to wait before funding an expensive approach and the innovation policy researcher trying to explain why some fields go decades without a breakthrough. It speaks directly to problems faced across the tech, energy, biotech and public policy sectors: Why do some markets look wide open but attract no good entrants? Why do investors sometimes reward expensive solutions nobody loves? Why does patent protection matter more in some domains than others? If you have ever asked ‘why has not someone solved this already?’ -- or been handed a budget to fund people trying to -- this lesson gives you the conceptual language to answer that question rigorously.

Goal: You will learn why the tech industry is shifting from ‘ideas are cheap, execution is everything’ back to recognizing that genuine ideas are actually scarce and valuable. You will understand the difference between having an idea and turning it into reality, and why society sometimes benefits from waiting for better ideas rather than jumping on the first one.

Real-World Applications:

Between 2022 and 2025, the cost of building and shipping software collapsed. AI coding assistants, no-code platforms and cloud infrastructure made execution cheaper than at any point in Silicon Valley's history -- yet the number of genuinely differentiated AI products remained surprisingly small and many well-funded teams converged on nearly identical applications. This is the idea scarcity thesis playing out in real time. Execution capacity flooded the market -- breakthrough concepts did not follow. Investors began penalizing ‘me-too’ startups not because the teams could not build, but because the ideas offered no conceptual moat. The lesson's framework maps directly onto this: when execution is cheap and abundant, the bottleneck shifts to the idea itself, the cost-to-wait calculus changes and the reward structure should -- and increasingly does -- tilt toward teams that can demonstrate a genuinely novel problem framing, not just faster delivery. 

The Problem and Its Relevance

For the past decade, Silicon Valley told everyone that ‘ideas are cheap -- execution is everything’. This meant the hard part was not thinking of an app or startup concept, but actually building it and getting users. If you had an idea for a new social media platform, investors would say your idea did not matter much because thousands of people probably had the same thought. What mattered was whether you could code it, market it and scale it.

But now something has changed. Leading AI researcher Ilya Sutskever recently said we have ‘more companies than ideas’ -- suggesting we have run out of good ideas even though we have plenty of people ready to execute them. Research backs this up: ideas for solving problems arrive randomly, to random people, at unpredictable times. You cannot just decide to have a breakthrough idea. This scarcity means ideas actually do have real value.

Here’s the challenge: when you get an idea, you face a ‘use it or lose it’ situation. If you do not pursue it, you will probably move on to other things and forget about it. But from society’s perspective, sometimes it is better to wait because someone else might come up with a cheaper or better way to solve the same problem. This creates a mismatch between what is best for you personally and what is best for everyone.

Why Does This Matter?

Understanding idea scarcity matters because:

(i) There is a trade-off between speed and cost: Imagine society is trying to solve a problem, like making better batteries for electric cars. If someone proposes an expensive solution today, we could implement it immediately but spend a lot of money. Or we could wait, hoping someone thinks of a cheaper solution -- but waiting means the problem does not get solved yet. Society has to balance ‘solve it now at high cost’ versus ‘wait for a better solution’.

(ii) Harder problems should offer bigger rewards: When ideas for solving a problem are really rare (like figuring out how to reverse aging or achieve nuclear fusion), we should be willing to accept expensive solutions because we might wait decades for a cheaper one. When ideas flow frequently (like new ways to organize a to-do list app), we can afford to be picky and wait for low-cost approaches.

(iii) Long delays signal scarcity: Patent law recognizes this through ‘long-felt need’. If a problem has existed for years without anyone solving it, that is evidence the problem is genuinely hard and ideas are scarce. An innovation that addresses a decades-old problem gets more credit than solving something that has been an issue for only a few months. Time-without-solution reveals how rare good ideas actually are.

(iv) You cannot plan when ideas will arrive: Ideas arrive randomly, like lightning strikes. You cannot schedule when you will have a breakthrough insight. Companies can hire smart people and give them resources, but they cannot guarantee those people will think of something valuable on a specific timeline. This randomness is why research is risky and why some teams can try for years without making progress.

(v) Individual and social incentives do not align: If you have an idea that costs $10 million to implement, you might pursue it because it is the best opportunity you have. But society might prefer you did not, hoping someone else will think of a $1 million solution next year. Since you will not personally benefit from someone else’s future idea, you have no reason to wait. This is why patents, prizes and funding decisions matter -- they try to align your incentives with what is best overall.

Three Critical Questions to Ask Yourself

• Do I understand the difference between having an idea (which happens randomly and unpredictably) and executing on it (which requires effort and resources but follows a more controllable process)?

• Can I explain why someone with an idea might be too eager to pursue it from society’s perspective, even though waiting feels irrational from their personal perspective?

• Am I able to give examples of problems where ideas have been scarce (long delays without solutions) versus problems where ideas flow frequently (many competing approaches)?

Roadmap

In pairs or small groups (10 minutes):

(i) Choose two problems: Pick one problem where solutions have been rare and slow to emerge (examples: curing cancer, achieving sustainable fusion energy, developing AGI, creating cheap desalination). Then pick one problem where many solutions exist or emerged quickly (examples: food delivery apps, social media features, project management software). For each, estimate roughly how long people have been trying to solve it.

(ii) Compare the scarcity: For your slow problem, count how many genuinely different approaches people have tried (not just minor variations). Estimate how many years passed between major new ideas. For your fast problem, do the same. Calculate a rough ‘ideas per year’ rate for each. Which one has scarcer ideas?

(iii) Discuss rewards: Should the slow problem offer bigger prizes or patents than the fast problem? Why or why not? Consider: if someone proposes an expensive solution to the slow problem, should society accept it or wait? How long would you wait before accepting an expensive approach? What information would help you decide?

(iv) Real-world check: Look up what actual rewards exist. Do Nobel Prizes, patents or venture funding in your slow domain actually offer bigger rewards than in your fast domain? If yes, does this match the theory? If no, why might real-world incentives differ from what theory suggests?

Individual Reflection

Write a brief response (5 minutes) to these prompts:

• Has this changed how you think about the phrase ‘ideas are cheap’? When might ideas actually be valuable versus when might they truly be abundant?

• If you had an idea for a startup or project, would you pursue it immediately or wait to see if you think of something better? What factors would influence your decision?

• Why might patent law give stronger protection to innovations that solve ‘long-felt needs’? Does this make sense given what you learned about idea scarcity?

Bottom Line

The key insight is simple: ideas and execution are different things. Ideas arrive unpredictably to random people -- you cannot control when you will have a breakthrough thought. Execution requires effort but follows a more predictable process. This distinction matters because it means ideas can be genuinely scarce resources, not just cheap starting points.

Society faces a fundamental trade-off: implement ideas quickly (accepting high costs) or wait for better ideas (accepting delay). When ideas are rare -- which you can tell by looking at how long problems go unsolved -- society should tolerate expensive solutions. When ideas flow frequently, society can afford to be selective. This is why patents on breakthroughs for decades-old problems often get broader protection than patents on incremental improvements.

What makes this tricky is that individuals face ‘use it or lose it’ constraints. If you do not pursue your idea, you will probably forget it or move on. But the next good idea will likely occur to someone else, not you. So from your perspective, your current idea looks more valuable than it might be from society’s perspective. Understanding this gap between personal and social value is crucial for thinking about innovation policy, startup decisions and research funding.

The tech industry’s shift from ‘ideas are cheap’ to ‘we have more companies than ideas’ reflects this economic reality. During the scaling era, execution capacity was the bottleneck -- anyone could think of an app, but few could build and scale it. Now, with abundant AI tools and engineering talent, the bottleneck has moved back to conceptual breakthroughs. Recognizing when ideas are scarce versus abundant helps you decide where to focus effort, how to evaluate opportunities and why some innovations receive more rewards than others.

#IdeaScarcity #IdeasVsExecution #Innovation #Creativity #LongFeltNeed #Entrepreneurship

{"@context":"https://schema.org","@type":"LearningResource","name":"Why Good Ideas Are Rare: Understanding Scarcity in the Age of AI","description":"A 15-minute lesson on why genuine ideas are scarce, how individual and social incentives diverge in idea markets, and what this means for innovation policy, startup decisions, and research funding.","educationalLevel":"Undergraduate","timeRequired":"PT15M","teaches":["idea scarcity","ideas vs execution","innovation economics","Poisson arrival of ideas","opportunity cost of early implementation","long-felt need doctrine in patent law","incentive misalignment between private and social value","R&D investment decision-making under uncertainty","startup ideation vs market validation","venture capital deal evaluation","intellectual property strategy","technology bottleneck analysis","first-mover vs fast-follower trade-offs","research funding prioritization","breakthrough innovation vs incremental improvement"],"keywords":["idea scarcity","ideas are cheap execution is everything","innovation economics","patent long-felt need","Ilya Sutskever more companies than ideas","R&D strategy","startup opportunity evaluation","opportunity cost","incentive alignment","venture funding","technology scarcity","breakthrough innovation","intellectual property","idea generation","innovation policy","entrepreneurship","AI era bottlenecks","conceptual breakthroughs","when to pursue an idea","cost-benefit of waiting for better solutions"],"dateModified":"2026-03-18","version":"1.0","versionNote":"Initial release — schema expanded March 2026 to include practitioner-facing teaches and keywords fields.","inLanguage":"en","learningResourceType":"Lesson","audience":{"@type":"EducationalAudience","educationalRole":["student","entrepreneur","venture capitalist","product manager","innovation strategist","R&D lead","policy researcher"]},"isAccessibleForFree":true}