Four-agent LangChain loop runs 11 days and burns $47,000 while dashboards look healthy

A multi-agent research system managed by Teja Kusireddy's team generated a $47,000 invoice due to a recursive communication loop. Two agents exchanged thousands of messages over eleven days without producing useful output, while dashboards showed healthy activity and normal latency -- the billing damage was surfaced only after the invoice arrived. The post-mortem identified two root causes: no per-agent budget ceiling, and no enforcement mechanism that would have terminated the session before the next API call.

What this means for youIf you run AI agents that call paid services, a loop can quietly run up a large bill for days before anyone notices. At work, set a hard spending limit per agent and alerts on billing, not only on dashboards showing activity. Check that your AI tools have a monthly spend cap and that someone gets an alert when usage spikes.