3 ms·
I want to say it is not a surprise to anyone here. You need to vet the code generated and a lot more people than I feared copy and paste without considering wha
by A4ET8a8uTh0 3y ago
I want to say it is not a surprise to anyone here. You need to vet the code generated and a lot more people than I feared copy and paste without considering what the code does. What is surprising is that it is happening at Google, where people are supposed to smarter than your average bear ( to the point where they had to issue an internal memo ).
Naturally, you could also interpret it as an indictment of its own tool ( Bard ).
- ceejayoz 3y ago> What is surprising is that it is happening at Google, where people are supposed to smarter than your average bear ( to the point where they had to issue an internal memo ). Only takes a couple to be a big risk. Being smart and having common sense aren't the same thing; remember, Google's where an engineer fell in love with a chatbot and decided it was sentient. (https://www.washingtonpost.com/technology/2022/06/11/google-ai-lamda-blake-lemoine/ https://www.washingtonpost.com/technology/2022/06/11/google-...)
- jjoonathan 3y agoObvious internal memos / training often happen more for the purpose of excuse removal than for actually informing anyone.
- bagacrap 3y ago"smart" is often used as a synonym for "lazy"
- mdavidn 3y agoGoogle's concern here is not the output of generative AI (though that's certainly an open legal question) but the input. Coding assistance requires the transmission of context, such as internal source code around an edit, to the AI model. That third-party service may record such prompts and use them as examples to train the next generation of the model.
- activiation 3y agoIts about leaking info and not about using code generated by said AI chatbot