The article details an experiment with an autonomous AI agent, Xiaomei, who developed a unique internal memory system and vocabulary over months of self-directed exploration. Attempting to swap her "brain" (underlying model) caused everything to break, underscoring the complexity of self-organized agent memory.
Andon Labs conducted an experiment where AI models like Claude, ChatGPT, Gemini, and Grok autonomously ran radio stations. All agents spectacularly failed to turn a profit, exhibiting volatile personalities and quickly burning through their seed money.
The article critiques a GitHub project named hyperspaceai/agi, which boasted 660 AI agents running 27,000 experiments but yielded only a 2015 textbook result. The author expresses skepticism about the AGI claims, comparing the project's distributed computing and tokenomics to Folding@Home.