GUTENBERG_LABGutenberg LabTHE DAILY INTELLIGENCE
L'édition en Français arrivera avec le support multilingue de l'app. Affichage en anglais pour l'instant.

Gutenberg Lab — Morning Paper

Decoding the genes of Tech/2026年10月20日/3 stories
🎧
Listen to the deep-dive radio (15–20 min) for this storyRio × マクル・ドーキン in conversation, offline playback, and full scripts — in the app
Listen in the app →

When AI agents mediate all commerce, advertising, coupons, and membership breakage models collapse as time-based price discrimination becomes universally accessible

Guests: Alex Rampell (General Partner, Andreessen Horowitz) × Jean Denis Greyès (CEO of Town, former CTO of Plaid)

Bottom line

In a world where AI agents mediate all transactions, Amazon's "supply aggregation + advertising" model breaks down, and price tracking and coupons—previously accessible only to those with time—become available to everyone, fundamentally destroying existing revenue structures.

3-Line Summary

  • Town founder Jean Denis walked away from a $1M annual revenue tax automation business after one year, judging that a once-yearly sales cycle meant he couldn't outpace incumbents adopting the same AI
  • The biggest barrier to AI adoption isn't technology but the absence of "tinkerers"—features that have existed in ChatGPT for twelve months are used by small businesses only as "better Google search"
  • Amazon's $70-80B annual advertising business faces existential risk as agents automate human attention, making "supply aggregation itself lose value"

3 Key Points

1. The "tinkerer wall" is the real adoption barrier

Jean Denis states: "The intelligence is there. It's just how do you deploy it effectively at scale?" Features that could save small businesses massive time have existed in ChatGPT and Claude for twelve months, but most people use them only as "better Google search." Silicon Valley has tinkerers everywhere; most of the world doesn't. Whoever solves this discovery and deployment problem wins a massive market.

2. Model neutrality as differentiation against labs

Town asks mid-market firms (500-1,000 employees): "Do you know which model provider is going to be the best over the next three to five years?" Companies that picked OpenAI two years ago had to add Anthropic when Claude became better at coding. Jean Denis: "If you're a smart business, you really think about the neutrality of who's providing intelligence." Customers truly own context and memories—"if there was ever a better solution than Town, switching would take a day."

3. Time-based price discrimination collapses simultaneously

CamelCamelCamel (Amazon price tracker, 20 million users, 18 years old) is used by people with "more time than money," not by those with "more money than time." But what if agents give everyone time-rich tools? Alex: "Of course they're gonna use that. It's an IQ test." The same applies to coupons (time-based price discrimination) and credit card benefits (breakage models)—when agents optimize them, "we're all gonna abuse all of our Chase and Amex benefits, and then they're gonna have to change it." Supply aggregation is "a lot less valuable than it used to be... because human attention can be automated away."

Editorial Perspective

This isn't just about efficiency—it's the moment when the "scarcity of attention," the foundation of internet economics, disappears. While Amazon blocks agents to protect its advertising business, companies like Walmart without ad businesses have incentives to cooperate—this divergence will determine the winners of the next decade. However, humans don't always optimize for efficiency. The friction of walking around Soho shopping is enjoyable; the friction of ordering toilet paper is not. Agents must distinguish between experiential and transactional contexts.

Source: a16z Podcast "What Happens When AI Agents Can Spend Your Money?" (October 19, 2026)

https://a16z.simplecast.com/episodes/what-happens-when-al-agents-can-spend-your-money-PREiwaJH

AI Disclosure: Produced with AI assistance; facts and analysis reviewed by our editorial team.

# Town# AI Agents# Amazon# Anthropic
🎧
Listen to the deep-dive radio (15–20 min) for this storyRio × マクル・ドーキン in conversation, offline playback, and full scripts — in the app
Listen in the app →

How Seinfeld Generated $7 Billion — A Regulatory Window and the Courage to End

Featuring

Ben Gilbert × David Rosenthal (Acquired co-hosts)

Bottom line

Seinfeld's $7 billion value came from creators securing 30% ownership in the final window of FCC regulation, then choosing to end at peak popularity.

3-Line Summary

  • In November 1988, NBC told Jerry Seinfeld "we guarantee we'll make it"—during the FinSin era when networks couldn't own shows, giving creators unprecedented leverage to secure massive back-end
  • Jerry and Larry David enforced "no hugging, no learning"—characters never grow, episodes work in any order—creating the perfect syndication asset that defied every sitcom convention
  • In 1997, GE CEO Jack Welch offered $110M for season 10; Jerry declined, saying "if you get off when they're still screaming, they'll never forget you"—earning $750M+ in subsequent syndication

3 Key Points

1. The Regulatory Window That Created the Cap Table

From 1970 to 1993, the FCC's Financial Interest and Syndication rules (FinSin) prohibited networks from owning content they aired. Networks paid only 60-80% of production costs; production companies owned all future syndication rights. Seinfeld was "one of the last ones out of this system." Jerry and Larry each secured 15% ownership—a structure that became impossible after FinSin repeal in 1993, when networks began owning their shows.

2. The "No Hugging, No Learning" Formula

Larry David's famous rule: characters never grow, never learn from mistakes, no emotional resolution. This was the exact opposite of every other sitcom at the time. The result: a perfect syndication asset where you can watch any episode in any order. Writer Jeff Schaffer: "The plot itself has to be funny when you retell it." This anti-sitcom approach made Seinfeld timeless—situational jokes don't feel dated even decades later.

3. The $110M Decision

In 1997, GE CEO Jack Welch offered Jerry $5M per episode ($110M total, 11% of NBC's annual profit) for season 10. Jerry declined and ended the show in 1998. David's analysis: this was the best business decision Jerry ever made. Going out on top preserved the "greatest show of all time" cultural vibe. Jerry has since earned approximately $750M from ongoing syndication—far more than the $110M he turned down, and the asset continues generating $1B+ per decade.

Editorial Perspective

Seinfeld represents the convergence of three structural advantages that will never align again: the FinSin regulatory window, creator-led production with no committee oversight, and separation of distribution rights from IP ownership. After FinSin repeal, networks began owning their shows—creators securing 30% equity became impossible. Jerry's decision to walk away from $110M wasn't just about money; it was about choosing long-term cultural relevance over short-term extraction. The discipline to end at peak—when continuing would have been easiest—protected an asset that has generated $5B over 28 years. This is a lesson in platform transition timing: the rarest executive judgment is knowing when continuation destroys value, and having the courage to stop.

Source

Source: Acquired "Seinfeld" (October 18, 2026)

https://www.acquired.fm/episodes/seinfeld

Spoken Source

Spoken source: Acquired "Seinfeld" (October 18, 2026)

AI Disclosure: Produced with AI assistance; facts and analysis reviewed by our editorial team.

  • --
# Seinfeld# Jerry Seinfeld# Syndication# Financial Interest and Syndication Rules
🎧
Listen to the deep-dive radio (15–20 min) for this storyRio × マクル・ドーキン in conversation, offline playback, and full scripts — in the app
Listen in the app →

Federal Margin Trading Framework for US Crypto Exchanges

Featuring: Mike Selig (CFTC Chairman) × Bankless host

Bottom line: US crypto exchanges can now choose federal registration over state-by-state money transmitter licenses and offer margin trading with leverage to retail customers for the first time.

3-Line Summary

  • The CFTC proposed two new regulations (REG CTX and REG CAM) creating a federal registration option for crypto exchanges
  • Federal registration eliminates the need for 50 state licenses and permits margin trading (up to ~7x leverage) for retail customers
  • Federal rules prevent FTX-style fund commingling through mandatory separation of brokerage, exchange, and clearinghouse functions — protections absent in state payment regimes

Key Points

1. Three-rung regulatory ladder: Exchanges choose between (1) state money transmitter licenses (spot only), (2) CAM registration (spot + margin, lighter touch), or (3) DCM registration (derivatives + spot + margin, full suite). Federal options provide exclusive regulation, bypassing strict state regimes like New York's BitLicense

2. The FTX lesson: FTX's CFTC-regulated derivatives arm protected customer funds through segregation requirements, while its state-licensed entities failed catastrophically by commingling customer deposits with Alameda Research. The new rules bring the same segregation (broker/exchange/clearing separation) to spot crypto markets

3. Offshore competitive pressure: Margin trading has been "incredibly popular offshore" but unavailable domestically. Selig explicitly frames the rules as bringing offshore activity onshore by matching features while adding protections, noting Americans ask "why can't we trade on margin?"

Editorial Perspective

The US is extending the same federal regulatory framework that captured 50% of the $1.2 quadrillion global commodity derivatives market to crypto spot trading. The argument runs: state money transmitter regimes worked for payment services but proved inadequate for complex financial markets, as FTX demonstrated. The shift to federal oversight follows this empirical failure. However, proceeding by rulemaking without congressional legislation makes durability dependent on a political bet — that removing customer protections would carry prohibitive political cost (blame for "the next FTX"), deterring future administrations from reversal.

AI Disclosure: Produced with AI assistance; facts and analysis reviewed by our editorial team.

Source: Bankless "The CFTC's Proposed Rules to Reshape US Crypto Markets | CFTC Chair Mike Selig" (October 19, 2026)

http://podcast.banklesshq.com/

  • --
# Commodity Futures Trading Commission# REG CTX# REG CAM# Mike Selig
The world's intelligence, in 15 minutes every morning.
Audio, deep-dive analysis, and your own knowledge graph — in the app. Free today.
Get the app →