OpenAI Codex token consumption surges unexplained

This sudden spike exposes the vulnerability of relying on large-scale AI agents without strict guardrails.

* Agentic runaway: As autonomous software agents gain wider adoption, small bugs or recursive task loops can trigger massive, unchecked token consumption spikes.

* Telemetry and transparency: OpenAI must improve user visibility into token billing and consumption, as unexpected surges create massive friction for enterprise developers.

* Infrastructure capacity: Sudden surges in Codex usage place severe stress on OpenAI's backend, necessitating robust resource allocation and prompt limit resets for affected accounts.