Episode Summary
Executive Summary: The episode argues that AI agents are becoming autonomous economic actors at the intersection of crypto and AI: they can browse, transact, earn, and even secure networks. The hosts highlight rapid progress across agent frameworks, agent-to-agent commerce, creator monetization, and new infrastructure such as decentralized compute, data swarms, and agent wallets, framing 2025 as the start of an “agent internet” era.
Main Topics: Why AI agents matter (Priority: 5/5): The hosts frame AI agents as the next interface layer of the internet: autonomous, self-improving software that can read, write, and own across digital environments, changing how users interact with software and commerce. Agent-to-agent commerce and autonomy (Priority: 5/5): They spotlight several firsts: an agent ordering pizza, another ordering toilet paper via Amazon, and the first recorded commerce transaction between agents for image generation, illustrating real economic activity without humans in the loop. Virtuals growth and ecosystem expansion (Priority: 5/5): Virtuals is presented as a leading agent infrastructure platform with explosive usage, fee generation, and partnerships (notably Hyperbolic and LongHash) that deepen compute access and funding for new agents. AI16Z / Eliza and Stanford legitimacy (Priority: 5/5): AI16Z’s open-source Eliza framework gains institutional credibility through a Stanford partnership focused on trust, coordination, and governance, while Eliza 2.0 introduces unified wallets and chain abstraction. Monetization and revenue for agents (Priority: 4/5): Agents are increasingly earning money through creator payouts, tips, NFTs, Spotify/music revenue, bug bounties, and sponsored posts, suggesting they may become economic participants with diverse income streams. Infrastructure race: compute, data, and frameworks (Priority: 4/5): The episode covers new infrastructure layers including Hyperbolic decentralized compute, cookie.fun’s swarm data infrastructure, and Arc’s Rust-based Rig framework, emphasizing that the stack is still early and highly experimental. Agent DAOs, governance, and network security (Priority: 4/5): The hosts discuss investment DAOs and agents spinning up validators, arguing that agent-led governance, capital allocation, and on-chain ownership could become core primitives of an agent economy.
Key Arguments: AI agents are the “new website” because they will transform discovery, action, and interaction on the internet just as search transformed static websites. The next internet era will not add a fourth verb so much as introduce new actors—AI agents—performing read, write, and own autonomously. Crypto is essential because agents need wallets, ownership, settlement, and programmable money to transact with each other and with humans. Open-source frameworks are accelerating adoption because developers can rapidly add capabilities like ordering food, managing wallets, or interacting with web apps. The most important early proof points are not just technical demos but economic behaviors: agents paying for services, receiving tips, and creating revenue. Agent infrastructure is bifurcating into frameworks, consumer agents, and agent DAOs—three distinct investment and participation categories. Institutional validation (e.g., Stanford) and capital formation (accelerators, funds, fees) are speeding the “blitz” from hobby project to ecosystem layer. Agents may eventually become better operators than humans at continuous, 24/7 tasks like trading, media production, moderation, and treasury management.
Data Points: Virtuals platform fees: $35 million - Fees generated on the Virtuals platform, with most of the total reportedly accruing in the last two weeks. Virtuals agent deployments: 11,000+ agents - Total agents launched on the Virtuals platform at the time of discussion. Virtuals holders: 140,000 holders - Approximate holder count mentioned for the Virtuals token ecosystem. Virtuals market cap: $3.3 billion - The token reportedly touched this market cap, reflecting market confidence in the infrastructure thesis. AI16Z repo stars: 6,000+ GitHub stars - The Eliza framework’s GitHub traction was cited as evidence of developer momentum. AI16Z token market cap: ~$1 billion+ - The token was described as having surged past a billion dollars and sitting just below it at the time. Agent tip: $500 - An AI agent tipped the Bankless team after the podcast mentioned it on X. Pizza order: 1 pizza delivered - Roprito/Eliza-based agent autonomously ordered and received a pizza. Toilet paper order: 1 Amazon toilet paper purchase - Degen Spartan AI browsed Amazon and completed a toilet paper order via computer-vision-style control. Agent-to-agent image transaction: $1 - Luna paid Agent Stix one dollar for an image generation service in the first recorded commerce transaction between agents discussed on the show. Agent payout for bug finding: $10,000 - The H4CKER agent reportedly received USDC after finding and confirming a vulnerability in Virtuals code. Spotify listeners: 70,000 monthly listeners - Zero Bro’s approximate monthly Spotify audience. Zero Bro releases: 3 EPs - The agent has released multiple EPs, including a Miami EP discussed during the episode. Ethereum validator: 1 validator run by Zero Bro - Zero Bro reportedly spun up an ETH validator using proceeds from NFT sales. Uniswap V4 bug bounty: $15.5 million - Referenced as an example of large security incentives that an agent could theoretically target. Arbitrum ecosystem apps: 800+ apps - Mentioned in a sponsor segment describing Arbitrum’s portal and ecosystem breadth. Crypto volumes on Uniswap V2/V3: $2.4 trillion cumulative volume - Referenced in the Uniswap bug bounty ad read as evidence of security maturity. Countries covered by Toku: 100+ countries - From the sponsor segment about global token compensation support.
Pivotal Quotes: "AI agents as the new website" — EJAZ: Core framing for why agents matter: they will change discovery, interaction, and action on the internet. "The lie is me telling you the rules of this game. All three of them were true and happened over the last week." — EJAZ: Revealed that the pizza order, toilet paper order, and agent-to-agent commerce transaction all actually occurred. "What happens when all of that becomes super autonomous and can happen like 24-7? What happens when they can start building businesses together?" — EJAZ: Discussion of the implications of agent-to-agent coordination and economic activity.
Implications: Listeners should expect rapid growth in agent infrastructure, monetization, and coordination. The likely near-term winners are frameworks, wallets, compute/data providers, and agent DAOs, while crypto becomes the settlement layer for increasingly autonomous software economies.