OpenAI's Internal AI Details Were Compromised Last Year, As Per a NYT Report