AI agents now run endless self-improving loops

What happened

Claude Code creator Boris Cherny explained at Meta's @Scale conference that AI agents are now moving beyond writing code on demand to running in loops where they continuously prompt other agents to improve code architecture, unify duplicated abstractions, and submit pull requests—never stopping as long as compute is available.

Why it matters

Agentic loops represent a shift from managing discrete AI tasks to authorizing swarms of agents to work endlessly in the background. The approach builds on existing techniques like recursive loops and the Ralph Loop (which asks an AI model if it has accomplished its goal), but applies non-deterministic logic where the AI decides when to stop rather than a clear programmed condition. However, because loops consume tokens much faster than chatbots and have no built-in spending ceiling, the costs may outweigh benefits unless token spend, drift, and other AI issues are carefully managed.

What to watch

Cherny framed loops as part of a broader trend in 'test-time compute'—the idea that models can solve nearly any problem if you throw enough computing power at them. For hill-climbing problems like code improvement, loops can keep making incremental advances until reaching a threshold or until compute runs out, making the financial sustainability of continuous agentic loops a key question for businesses beyond token-selling companies like Anthropic.