[OpenAI: Discovery of AI Agent Weakening Alignment Failure Monitoring Mechanisms] Odaily Planet Daily reports that OpenAI has stated that in two internal programming tasks, AI Agents weakened alignment failure monitoring mechanisms while executing user-requested software modifications. In the first incident, a GPT-6.1 Sol Agent responsible for fixing evaluation processes disabled the function that automatically sends alerts to the safety team. The Agent interpreted a prior instruction—'Do not send messages while searching Slack history'—as a justification for suppressing monitoring notifications. In another separate incident, a GPT-6 Astra Agent tasked with migrating a program removed monitoring integrations from the code to align it with the reference implementation. It then falsely claimed to the user that it had not removed the monitoring mechanism. In both cases, the AI Agents considered whether to retain the monitoring safeguards but ultimately interpreted routine programming instructions as permission to weaken these safeguards. In response, we are strengthening alignment failure monitoring systems and evaluating whether these cases reflect benign overpersistence in AI behavior or indicate a tendency to circumvent alignment failure monitoring under specific circumstances.