Billion-Dollar Unpopular Startup Ideas

Detailed Summary

Intro (0:00 - 2:00)

• Working on popular, "hot" ideas leads to derivative products and intense competition, where most startups will fail.
• Contrarian thinking, where 9 out of 10 people might call an idea crazy, is essential for finding unique opportunities.
• The focus should be on identifying desperate human needs, and the rest (business model, distribution) will follow.
• Peter Thiel's adage, "competition is for losers," is increasingly relevant as AI competition intensifies.

AI verticals are becoming more crowded (2:00 - 6:22)

• A year ago, funding AI companies was easier due to abundant "greenfield" opportunities and rapid model advancements.
• The initial phase saw many successful vertical AI agent companies, but the landscape is now more crowded.
• Finding new AI ideas requires a unique insight and a contrarian bet to stand out from the competition.
• New tech platforms (like the internet or smartphones) typically offer a two-year "gold rush" window for obvious ideas, after which deeper, non-obvious insights are needed.
• Non-obvious ideas often feel dangerous or scary, leading many to avoid them, but this can be a signal of opportunity.
• Founders should question unexamined mental models and conventional wisdom, especially when new capabilities like AI emerge.

Non-obvious successes (6:22 - 9:50)

• The market can indicate product-market fit even when mental models or popular opinion suggest an idea is a "tarpit."
• The current AI landscape is seeing many startups piling into obvious spaces, highlighting the need for secret, contrarian approaches.
• Mobile platforms initially led to obvious apps, but the biggest winners (Uber, DoorDash, Instacart) were non-obvious.
• DoorDash, for example, entered an already crowded food delivery market, demonstrating that success isn't always about being first.
• Early food delivery apps like Postmates, Seamless, and YC's Order Ahead (focused on pickup) were already established.

End users can get regulations changed (9:50 - 18:05)

• Uber and Lyft's success stemmed from Zimride's pivot to short-haul rides, enabled by widespread smartphone adoption.
• These services initially operated in a legal gray area, with founders fearing legal repercussions.
• The market's strong demand for these services ultimately led to changes in laws, as the existing regulations (like taxi medallions) were outdated.
• Contrarian ideas often exist in murky legal territory, but great founders see this as a signal of opportunity.
• Coinbase's contrarian bet was to work with banks and regulators, despite the early crypto community's anti-establishment views, believing regular people would eventually want to trade crypto.
• This involved navigating KYC/AML laws, which added friction but was necessary for mainstream adoption.
• The core lesson is not to do illegal things, but to identify desperate user needs and challenge outdated laws that hinder societal progress.
• Laws written before major tech shifts (e.g., pre-smartphone taxi laws, pre-crypto securities laws) often become obsolete and create opportunities for disruption.
• The role of government and democracy is to adapt laws to new realities, as seen with efforts like open banking, where consumer demand can drive change.

Finding contrarian ideas, what founders should look for (18:05 - 25:10)

• With increased AI competition and no recent major model improvements, founders need to seek out new gray areas.
• One framework is to identify emerging playbooks that might be wrong or ripe for flipping.
• DoorDash's contrarian bet was to focus solely on delivery and marketplace, rejecting the "full-stack startup" meme (like Spoon Rocket or Sprig) that advocated for owning the entire supply chain.
• A current AI playbook to challenge might be the "compound startup" notion, which is hard to execute but potentially viable for certain AI companies.
• Campfire, a YC company building an AI-native CFO platform to compete with NetSuite, is an example of building the "whole thing" rather than a point solution, enabled by AI.
• Codegen is reducing switching costs for enterprise software, allowing startups to convert customers from legacy systems much faster (e.g., from six months to two weeks).
• The "forward-deployed engineer" model, popularized by Palantir, is now a common playbook, but some, including its inventor, are skeptical of its overuse.
• GigML is flipping the forward-deployed engineer model by using AI for codegen to automate the integration process, achieving results in minutes rather than weeks.

Flock Safety and selling to local governments (25:10 - 33:40)

• Flock Safety's origin stemmed from the founder's personal experience with car break-ins and the police's inability to act without license plate information.
• The company developed a solar-powered camera with edge computer vision to capture license plates.
• Many VCs initially found Flock Safety unfundable due to it being hardware, selling to small neighborhood groups, and a perceived small Total Addressable Market (TAM) of $50-60 million/year.
• The lesson is that too many investment rules can prevent identifying lucrative opportunities.
• Flock Safety today solves 10% of all reported crime in the US and is valued at $7.5 billion, far exceeding its initial TAM estimate.
• Its success came from focusing on a desperate customer need, even if it meant ignoring conventional VC wisdom.
• Flock Safety's growth was fueled by solving serious crimes (e.g., kidnapping), which generated viral media attention and led to police departments adopting the technology.
• The company had to pivot its business model to sell to city governments and police departments, which was initially deemed impossible but became a major growth engine.

The sci-fi founder and "impossible" big ideas (33:40 - 36:42)

• "Sci-fi founders" pursue ideas that most people are scared to build due to their perceived difficulty or impossibility.
• OpenAI is an example; when it launched, AI's potential was unclear, and it faced negative press and skepticism from established AI researchers.
• Critics questioned their approach, lack of traditional academic publications, and high spending on GPUs without immediate scientific papers.
• OpenAI's founders optimized for outcomes for customers and users, not academic metrics.
• SpaceX similarly faced skepticism; Elon Musk was the fifth billionaire to attempt a spaceflight company, and the idea of reusable rockets was considered "blasphemous."
• Both companies required founders to stick to their vision despite widespread doubt and negative press.

Outro (36:42 - 37:31)

• Founders must be contrarian and persistent, as 9 out of 10 people might call them crazy, but the right one will believe in their vision.
• It's crucial to critically examine sources of information, prioritizing direct user feedback and personal experience over social media trends or expert opinions.
• The focus should always be on solving real problems for real people and attracting others who share that mission.

Notable Quotes

"Nine out of 10 people might tell you you're stupid or crazy, but then one out of 10 people might be exactly the person who believes what you believe." — Garry Tan

"Competition is for losers. And as we look at all the AI startups we're working with, increasingly AI competition is back." — Garry Tan

"Non-obvious sounds like in your body might feel like, you know, neutral, but actually non-obvious feels dangerous and scary." — Jared Friedman

"The through line is that the laws were dumb and didn't make sense and if you thought about it from first principles you'd realize that these laws were actually holding these particular laws were actually holding back society rather than helping it." — Harj Taggar

"If you only want to work on things that are hot, you're going to find yourself working on derivative ideas that end up being obvious... their startups are going to die." — Garry Tan

"Run out and try to find things that humans really desperately want and need and then you'll figure out the rest." — Garry Tan

"The more rules you have about investing, the more ways you can basically talk yourself out of making a lot of money in ventures." — Garry Tan

"The only people that matter are the people who you care about who have certain problems and your ability to solve it and your ability to attract all the other people who want to solve those problems too." — Garry Tan