AI proof businesses worth building in 2027

My First Million My First Million • Nov 04, 2026

Audio Brief

Show transcript
Referencing entrepreneur Pieter Levels, Sam Parr notes that artificial intelligence has wiped out technology as a durable business moat, leaving communities, networks, and marketplaces as the best defenses left. In their episode released on November 4, 2026, hosts Sam Parr and Shaan Puri break down how these human networks and consumer psychology drive massive businesses. By the end of this summary, you will know how to apply peer groups, big spenders, and mystery box mechanics to build a lucrative company. You can build an AI-proof business by connecting non-competing peers. Since the 1970s, the National Automobile Dealers Association has charged car dealers six thousand to eight thousand dollars a year to join groups of twenty-four non-competing peers from different regions to audit each other's profit and loss statements line by line, a program Parr guesses makes thirteen million to fifteen million dollars annually. Another lucrative model organizes local business owners into structured referral networks. Business Network International groups roughly thirty non-competing local professionals, like a realtor, a lawyer, and a landscaper, to meet weekly and share leads, with its website citing over three hundred fifty-five thousand members generating nearly twenty-seven billion dollars in business over twelve months. That same human trust can power multi-million-dollar luxury marketplaces on free platforms. Moda Clubs runs dozens of Facebook communities and chatrooms where over six hundred thousand members trade watches, cars, and jewelry, and vetted dealers pay up to ten thousand dollars a year to list items in a business Parr estimates makes five million to ten million dollars a year. Relying on a tiny fraction of heavy spenders, known as whales, is also the secret behind free-to-play gaming and streaming. Puri explains that across Candy Crush's eighty-two million monthly active users, one hundred percent of the revenue comes from just four percent of players. Digital card apps take this psychology further by combining Japanese mystery capsule mechanics with instant buybacks. On apps like Triumph, a user can pay fifty dollars to digitally open a pack of collectible cards, take an instant buyback for forty-three dollars in store credit on a mediocre card, and leave Triumph the seven-dollar difference risk-free without ever shipping the physical card. Solving a private problem with a simple tool can also build a quiet empire. Parr highlights Private Photo Vault, an app with one million App Store ratings that hides private photos behind a lock and previously disguised itself as a calculator. Even the largest digital networks face massive security risks, pushing users toward automated alternatives. Puri discusses a massive crypto exchange hack at Binance, calling it a five-billion-dollar theft, though an Associated Press article states over six billion dollars was taken, and hedges that according to the last thing he read, North Korea executed the heist and laundered the funds through a permissionless decentralized finance protocol. As artificial intelligence makes software easy to copy, human networks and timeless consumer psychology remain durable ways to build a business. Try applying that psychology to physical retail, where data shared with Puri by a toy industry friend showed that taking toys off open shelves at Walmart and Target and putting them inside blind mystery boxes generated seven times the revenue.

Episode Overview

  • Hosts Sam Parr and Shaan Puri explore how entrepreneurs can build lucrative, AI-proof businesses by leveraging peer communities, referral networks, curated marketplaces, and consumer gambling psychology.
  • The conversation begins with high-margin, low-tech community models for traditional industries, looking at how organizations like the NADA 20 Group and Business Network International generate massive recurring revenue.
  • The hosts then examine how private Facebook groups can evolve into multi-million-dollar luxury marketplaces built on vetted trust and funded by a tiny fraction of high-spending users.
  • Next, they analyze the economics of "whales" and "gacha" mystery mechanics across mobile gaming, physical retail toys, and digital collectible card apps generating up to $1 billion in transaction volume.
  • The episode concludes with a look at a discreet, high-revenue utility app and a discussion on cryptocurrency security, decentralized finance, and a multi-billion-dollar exchange hack.

Key Concepts

  • AI-Proof Moats via Non-Competing Peer Communities: Referencing Pieter Levels, Sam Parr notes that technology is no longer a durable business moat because of AI, leaving communities, networks, and marketplaces as the best remaining defenses. Parr (who operates the founder community Hampton and aims to build a multi-billion-dollar company generating nine figures in free cash flow by acquiring communities) and Shaan Puri (who previously ran the e-commerce community Club LTV) highlight the National Automobile Dealers Association (NADA), launched in 1917. Its "NADA 20 Group," started in the 1970s, charges auto dealers $6,000 to $8,000 annually to join groups of 24 non-competing peers from different regions who fly in three times a year to audit each other's Profit and Loss (P&L) statements line by line. The program has expanded to HVAC companies, marinas, and boat repair shops, and Parr guesses it generates $13 million to $15 million a year in revenue. Similarly, Puri highlights "The White Coat Investor" (founded in 2011), a blog and Facebook group community where doctors and dentists discuss personal finance and rental property cap rates.
  • Structured Local Referral Networks: Business Network International (BNI) groups roughly 30 non-competing local business owners (such as a realtor, accountant, HVAC operator, lawyer, landscaper, and garage door service) to meet weekly at venues like a YMCA to share leads. While BNI's website data as of December 31, 2025, cites 355K+ global members across 11,800+ chapters in 78 countries generating 17.8 million referrals and $26.9 billion in business over the prior 12 months, Parr verbally claims the group has 800,000 members, costs low-digit thousands of dollars a year to join, and was rumored to have been recently acquired for over $1 billion.
  • Trust-Based Niche Marketplaces: Founded in 2018 by a man in Sioux Falls, South Dakota, Moda Clubs operates 23 Facebook communities and 39 chatrooms with over 600,000 members trading watches, collectible cars, gold, and jewelry. Watch groups are segmented into tiers under $10,000, above $10,000, and above $100,000. Buyers join for free, while dealers must be vouched for by an existing member, pass a vetting interview with the founder, and pay $1,000, $5,000, or $10,000 annually to list items. The founder told Parr via direct message that he employs over a dozen full-time staff and that the top 1% or 0.001% of users pay to keep everything running; Parr estimates the business makes $5 million to $10 million in annual revenue.
  • Whale Economics in Gaming and Streaming: In free-to-play ecosystems, a tiny fraction of users ("whales") subsidize the entire user base. Puri (who formerly worked at Twitch) explains that in mobile games like Candy Crush, 80% to 90% or all of the revenue comes from whales spending thousands of dollars a year, while free players and those paying $2.99 to remove ads make up the rest; specifically, across Candy Crush's 82 million monthly active users, 100% of revenue comes from 4% of players. At Twitch, launching a "gifting subs" feature allowed whales to buy 100 or 200 subscriptions at $5 to $6 per month each (costing around $1,000) for random viewers, driving up channel revenue by giving whales a way to buy social status like bottle service at a club. Puri notes his Bebo co-founder Furqan (formerly CTO of mobile game ad network AppLovin) observed that the exact same whales spend heavily across Candy Crush, Block Blast, and FarmVille. Unlike Las Vegas slot machines that are legally required to maintain an ~89% payout rate, mobile game developers face no rules and scientifically engineer game economies. Puri adds that a content creator analyzing app source code claimed mobile games track players who paid previously but haven't spent in 30 days, feeding them a winning streak followed by a frustrating wall to trigger a purchase, though Puri hedges that the code isn't public and his understanding might be slightly off.
  • Digital Card "Ripping" and Gacha Buyback Mechanics: Apps like Triumph, Collector Crypt, and Logan Paul's RipIt apply Japanese "Gacha" (mystery capsule toy) mechanics to digitally opening packs of collectible cards like Pokémon. Puri hedges that Gacha games became so popular in Japan they are now outlawed by law, and notes that unlike video game loot boxes in Overwatch or FIFA that only yield aesthetic digital skins, card-ripping apps attach real-world financial value (such as a top Pikachu card selling for $15 million) to the gambling mechanic. In Triumph (founded by a former Citadel and Bridgewater quant, according to Parr), a user can pay $50 to digitally open a card pack; if they get a mediocre card, the app offers an instant buyback, such as $43 in store credit. The company books $50 in GMV/revenue, keeps the $7 spread risk-free without shipping the physical card, and recycles the card back into the digital deck. Puri cites a dinner conversation indicating Triumph did $1 billion in revenue/GMV in the last year (compared to Snapchat's ~$4 billion), and notes that on-chain fee data shows blockchain-based Collector Crypt grew from $50,000 a month to $25 million a month, and is currently doing $100 million a month.
  • Discreet Consumer Utility Apps: Simple utility apps solving private problems can quietly become massive businesses. Parr highlights "Private Photo Vault - Pic Safe," an app with 1 million App Store ratings and reviews that hides private or nude photos behind a lock (and previously used a calculator icon). After meeting the Las Vegas-raised founder at a Hampton event and seeing his confidential employee and revenue figures, Parr confirmed the company is a long-running financial behemoth.
  • Crypto Security Risks and Permissionless DeFi: Puri (a crypto investor and enthusiast) notes there is talk that AI hacking and quantum computing could break encryption protocols, representing one of the few existential risks to Bitcoin alongside Satoshi Nakamoto waking up and moving coins. Discussing a massive Binance hack (which Puri calls a $5 billion hack, while an Associated Press article on screen dated Oct 30, 2026, states $6.3B US was hacked), the hosts explain that after a smaller 2018 hack, Binance allocated 10% of trading fees into its SAFU (Secure Asset Fund for Users) fund, growing it to ~$2 billion, which it is combining with its corporate treasury to make users whole. Puri hedges that according to the last thing he read, North Korea executed the hack by having a woman romantically date a Binance engineer to physically steal his laptop and drain the hot wallet. To launder the trackable funds before protocols froze them, the hackers used Tornado Cash, split funds across wallets, and deposited the stolen crypto into a permissionless DeFi protocol as collateral to borrow up to 50% of the value in clean stablecoins. Puri predicts overlooked DeFi protocols will make a massive comeback, similar to how crypto prediction markets like Augur failed in 2015 to 2017 before Polymarket and Kalshi broke out eight years later.
  • Episode Disclosure: The podcast is part of the HubSpot Media network and is sponsored by HubSpot.

Quotes

  • At 1:59 - "It's the most simple peer group ever. All you do is you bring a P&L statement of your business, of your dealership, and you sit down with 20 others and you go line for line and you explain your profit and loss statement and other people can look at you and being like, 'Oh, you're killing it, I'm going to copy that,' or, 'You suck, have you thought about doing this?'" - Sam Parr explaining the high-value, low-tech format of the NADA 20 Group for auto dealers.
  • At 10:38 - "Pieter Levels has been talking about like what moats are going to go away with AI, and previously technology was a moat - that's not a moat anymore. And I think communities, networks, and marketplaces are still a really good moat, and this is a really interesting business that this guy started on Facebook groups." - Sam Parr on why human networks like Moda Watch Club are resilient against artificial intelligence disruption.
  • At 13:20 - "Candy Crush has 82 million monthly actives and all the revenue comes from 4% of the players. So 96% of the players are just kind of the cost of doing business, just the cost of finding those whales." - Shaan Puri illustrating how free-to-play businesses rely entirely on a tiny fraction of heavy spenders.
  • At 17:34 - "And for the company, what they did was they booked - so they got 40, 50 bucks of revenue, they give 43 back, 7 is free. They never had to sell, to give you the card. They just put the card right back in the deck and shuffle, they'll sell that same card again to somebody else." - Shaan Puri breaking down how digital card-ripping apps use instant buybacks to generate risk-free spreads without shipping physical inventory.

Takeaways

  • Launch a high-ticket peer group for unglamorous, high-earning local industries (such as dental practices, law firms, HVAC owners, or marinas where owners net around $1 million a year) by placing non-competing owners from different geographic regions into groups of 20 to 24 to review their P&L statements together several times a year.
  • Design tiered pricing and status-signaling features for "whales" in your product or community, allowing the top 1% to 4% of heavy spenders (such as vetted luxury dealers or status-seeking community members) to subsidize a free experience for the remaining 96% to 99% of users.
  • Incorporate "mystery box" psychology into physical retail or e-commerce products to drive higher sales volume; as data shared with Puri by a toy industry friend from Walmart and Target demonstrated, taking the exact same toys off open shelves (where consumers have choice) and placing them inside blind mystery boxes (like LOL Surprise dolls or mystery squishy dumplings) generated 7x the revenue across five different product categories.