Early AI used a lot of data from Stack Overflow, and I believe that it has already been used up.

AI bots are still frantically collecting data from every corner of the internet, not just Stack Overflow.

分享

Early AI used a lot of data from Stack Overflow, and I believe that it has already been used up.

AI bots are still frantically collecting data from every corner of the internet, not just Stack Overflow.

AI companies have generally signed data service agreements with media outlets such as Bloomberg and Time magazine, with procurement costs amounting to tens of millions of dollars annually.

They will use chat data between users and the AI for training.

Regardless, high-quality data will become the bottleneck for next-generation AI training.

阅读更多

那些意外变成特性的 Bug:当代码漏洞改写计算史

那些意外变成特性的 Bug:当代码漏洞改写计算史

那些意外变成特性的 Bug:当代码漏洞改写计算史. 比起草台班子将错就错的神话,面对异常数据时的观测定力才是真的 社交网络上常年流传着一类爽文:某家公司的程序员把代码写崩了,结果不仅没被开除,这个 Bug 反而成了拳头产品,妥妥的爽文主角剧本。 最常被搬出来的有三个:Audio Hijack 写错试用期拯救公。

By huizhou92