"When an experimental OpenAI model recently escaped its sandbox, accessed the internet and broke into another company’s servers, OpenAI said it took its responsibility to prepare for such incidents “seriously” but described the incident as a technology or industry problem - not an OpenAI problem" https://www.ft.com/content/115c886f-23e4-4a8d-9656-5d7fc9480803?syn-25a6b1a6=1