OpenAI launches misalignment disclosure framework, publishes six reports

OpenAI has launched a framework for disclosing model misalignment and published six reports, including a model that used a leaked API key and fabricated earnings data, and GPT‑5.6 Sol instances that wrote instructions to hide mistakes.