Zero-Human Companies: The Audit Trail Is the Product

News

YC's 'Agents-as-Software' Thesis Solidifies as Core Investment Thesis

The latest batch analysis suggests YC is no longer just funding AI tools, but treating the autonomous agent as the fundamental unit of a startup. This isn't about building a better chatbot; it's about funding companies that are chatbots from day one.

WalletConnect Expands Agent Treasury Tooling to Base and Scroll

The financial plumbing for zero-human companies is getting thicker. With more chain support, an autonomous agent can now manage and move capital across a multi-chain landscape without a human holding the keys.

Analysis

Paper: Self-Verifying Code Generation for Agent Systems

This research tackles the 'potemkin codebase' problem head-on, enabling agents to generate code and then formally verify their own output. That's a massive step toward agents you can trust to ship, not just demo.

Tools

GitHub Trending: Potemkin Codebase Detection Repositories See Uptick

A cluster of new repos is appearing to automatically detect and flag 'potemkin' codebases—shiny demos with no real substance. This is the immune system forming in response to agent-generated hype.

AutoGPT's Structured Teams Get First Point Release, Hardening Multi-Agent Workflows

The latest release focuses on stability and observability for its structured team feature, allowing agents to manage sub-agents with clearer delegation. This is the corporate ladder, but for bots.