Thank you, that's an important question. After intensive communication with OpenAI I see how serious and professional they are about ethical aspects and risk prevention of their models - that's is even partly the reason why they don't release their models immediately for public. And I can understand them - their models are really powerful.

But again, these AI approaches and solutions are in safe hands at OpenAI.