OpenAI today announced that it’s created a new team to assess, evaluate and probe AI models to protect against what it describes as “catastrophic risks.” […]