Stripe's John Collison on How Agentic Commerce Will Reshape the Internet
Stripe's John Collison on How Agentic Commerce Will Reshape the Internet
2 hours agoOdd LotsBloomberg
Podcast47 min 26 sec
Listen to Episode
Note: AI-generated summary based on third-party content. Not financial advice. Read more.
Quick Insights

Investors should look to Cloudflare (NET) as a primary beneficiary of the shift toward "agentic commerce," as its partnership with Stripe makes it the preferred infrastructure for AI agents to autonomously host websites and register domains. Consider Stripe (currently private) as the essential "toll booth" for the AI economy, specifically for its ability to capture high-volume microtransactions that were previously economically unviable. The rise of AI agents suggests a move away from traditional SEO-heavy retail toward niche brands, favoring platforms that provide machine-readable data for AI "shoppers." To capitalize on the "pay-by-the-sip" model of AI services, monitor the integration of Stablecoins as a low-cost settlement layer for automated B2B API calls and data queries. While the sector is growing, be mindful of risks regarding AI "innumeracy" in financial math and the need for businesses to replace bot-blocking CAPTCHAs with verified agent access.

Detailed Analysis

Stripe (Private)

• Stripe is positioning itself as the primary financial infrastructure for "agentic commerce"—a shift where AI agents, rather than humans, execute transactions. • The company recently announced a partnership with Cloudflare to allow AI agents to buy and host domain names autonomously, removing manual configuration hurdles. • John Collison (President/Co-founder) identifies two main categories of agentic commerce: * Consumer-led: AI researching products and completing the final checkout step to reduce friction. * B2B/Developer-led: AI agents (like Claude Code) autonomously purchasing resources like API keys, hosting, or data to complete a task. • Stripe is working with major AI players (OpenAI, Microsoft, Google, Meta) to build the "mechanical wiring" for these transactions.

Takeaways

Reduced Friction as a Growth Driver: Stripe’s strategy is based on the historical trend that the lowest-friction payment option always wins. By enabling "one-click" agentic checkouts, Stripe aims to capture a new class of automated transaction volume. • New Revenue Streams: The rise of AI agents makes microtransactions (e.g., paying $0.40 for a single data query or article) economically viable for the first time because the "mental load" of the decision is handled by the AI. • Infrastructure Play: Investors should view Stripe as a "toll booth" for the AI economy. As AI agents begin "vibe coding" and building apps, Stripe earns fees on the underlying resource purchases.


Cloudflare (NET)

• Cloudflare is collaborating with Stripe to enable seamless, automated domain registration and web hosting. • This partnership aims to solve the "maddening" manual process of toggling DNS fields and A/B records by allowing an AI agent to handle the backend setup via API.

Takeaways

Legibility to Agents: Cloudflare is moving toward making web infrastructure "machine-readable," which could make it the preferred provider for AI-driven development platforms. • B2B Efficiency: By removing the technical barriers to launching websites, Cloudflare may see an uptick in new domain registrations and hosting instances driven by automated tools.


AI & Agentic Commerce Themes

The Death of Keyword Search: The transcript suggests that traditional keyword search (e.g., typing "Nike shoes" into a box) is becoming obsolete. AI-powered research allows for "textual search interfaces" that understand complex constraints (e.g., "find a bike case that fits these specific dimensions"). • Impact on Small Brands: Agentic commerce may benefit smaller, niche merchants. AI agents "have good taste" and can discover high-quality, lesser-known brands that would otherwise be buried by the SEO dominance of large retailers. • Machine-Readable Retail: Businesses will likely need to optimize their websites for agents (AI SEO) by providing real-time data on stock levels, sizes, and pricing in a format LLMs can ingest.

Takeaways

Investment Shift: Traditional advertising-heavy companies may face challenges if AI agents bypass "scrolling" and go straight to the best-value product. However, Brand Affinity still matters as humans will likely remain the "final decider" in the loop for high-stakes purchases. • Economic Dynamism: Stripe reports a 71% year-over-year increase in new business creation on its platform, suggesting that AI is lowering the barrier to entry for entrepreneurs, potentially leading to a more fragmented and competitive retail landscape.


Stablecoins & Crypto

• Stablecoins are identified as a potential solution for the "pay-by-the-sip" model of the AI economy. • Because AI agents operate with marginal costs for every "token" used, traditional credit card fees are often too high for individual micro-tasks.

Takeaways

B2B Settlement: While not yet widely adopted, Stripe is building infrastructure for stablecoins to allow agents to pay for services (like data or API calls) without needing a pre-existing relationship or a credit card on file for every vendor. • Efficiency: Stablecoins could provide the "roaming" capability for the internet, allowing agents to move between services and pay tiny amounts instantly.


Risk Factors

Innumeracy of LLMs: A significant risk mentioned is that current AI models are "innumerate"—they are excellent at language and coding but frequently fail at basic financial math and spreadsheet logic. • Bot Protections: Many websites currently have "bot protections" (like CAPTCHAs) that hinder agentic commerce. Businesses will need to "unwind" these protections for verified agents (a "TSA Pre-check for bots"). • Liability and "Rogue" Bots: There are concerns regarding who is responsible if an AI agent makes a poor or unauthorized purchase. The suggested solution is "delegated authority" with strict spending limits, similar to corporate credit cards.

Ask about this postAnswers are grounded in this post's content.
Episode Description
The internet is made for shopping. For years, the main inputs for e-commerce transactions involved targeted ads, algorithmic recommendations, SEO, and lots of mindless scrolling. But agentic commerce might represent a sea change for e-commerce: With the rise of AI agents doing shopping on behalf of consumers, how are retailers going to adapt? John Collison, co-founder of the financial services and payment processing company Stripe, has first-hand experience with all the ways e-commerce has changed in the last decade, and he thinks agentic commerce is going to completely transform the online shopping experience. On this episode, we speak to Collison about how AI has already changed the way consumers make purchasing decisions, why keyword search is a "ridiculous" way to find things to buy, what it means when brands will have to appeal to AI agents as opposed to human buyers, and if AI agents can truly mimic human taste. Subscribe to the Odd Lots Newsletter Join the conversation: discord.gg/oddlots See omnystudio.com/listener for privacy information.
About Odd Lots
Odd Lots

Odd Lots

By Bloomberg

<p>Bloomberg's Joe Weisenthal and Tracy Alloway explore the most interesting topics in finance, markets and economics. Join the conversation every Monday and Thursday.</p>