Of course, you can't necessarily predict from a mouse or a dog what's going to happen in a human. And there have been a number of cases where substances that have proven to be toxic in animals have been proven to be safe in humans and vice versa. There are also of course animal welfare limitations to animal testing. Most people, I think, if they had the choice, would want their substances to be used on as few animals or no animals as possible while still ensuring safety. Now that's been a really difficult needle to thread, but AI might help to make that more possible.
當然,你不一定能從老鼠或狗身上預測到人類身上會發生什麼。已經有很多案例證明,在動物體內有毒的物質在人體內是安全的,反之亦然。當然,動物試驗也有動物福利方面的限制。我認為,如果可以選擇,大多數人都希望在確保安全的前提下,儘可能少地或不在動物身上使用他們的物質。現在看來,這真是一根很難穿的針,但人工智能或許能讓這更有可能實現。