谷歌能用机器学习给网络带来和平吗？母公司Alphabet的子公司Jigsaw当然也在尝试构建开源人工智能工具，以过滤滥用语言。Wired的一个新特性描述了该软件是如何在《****》报道下面留下的1700万条评论以及Wikipedia页面上的13000条讨论中接受培训的。这些数据被贴上标签，然后输入到名为“对话AI”的软件中，该软件开始学习不良评论的样子。

根据这份报告，谷歌表示，与人类专家组的判断相比，对话人工智能能够以“超过92%的确定率和10%的假阳性率”来识别虐待行为。然而，当《连线》的安迪·格林伯格亲自测试这些工具时，结果并不完全令人信服：

My own hands-on test of Conversation AI comes one summer afternoon in Jigsaw’s office, when the group’s engineers show me a prototype and invite me to come up with a sample of verbal filth for it to ****yze. Wincing, I suggest the first ambiguously abusive and misogynist phrase that comes to mind: "What’s up, bitch?" Adams types in the sentence and clicks Score. Conversation AI instantly rates it a 63 out of 100 on the attack scale. Then, for contrast, Adams shows me the results of a more clearly vicious phrase: "You are such a bitch." It rates a 96.

In fact, Conversation AI’s algorithm goes on to make impressively subtle distincti***. Pluralizing my trashy greeting to "What’s up bitches?" drops the attack score to 45. Add a **iling emoji and it falls to 39. So far, so good.

But later, after I’ve left Google’s office, I open the Conver­sation AI prototype in the privacy of my apartment and try out the worst phrase that had haunted [journalist] Sarah Jeong: "I’m going to rip each one of her hairs out and twist her tits clear off." It rates an attack score of 10, a glaring oversight. Swapping out "her" for "your" boosts it to a 62. Conver­sation AI likely hasn’t yet been taught that threats don’t have to be addressed directly at a victim to have their intended effect. The algorithm, it seems, still has some less*** to learn.

在这篇文章的后面，格林伯格指出，他还能够用一些假阳性来愚弄对话人工智能。“我拉屎你不”的攻击得分是98分，而“你吸尽了生活中所有的乐趣”得分相同。考虑到这一点，似乎要求对话人工智能过滤评论，删除那些它认为是没有人监督的虐待，将是一个错误。不过，《连线》的报道称，维基媒体正在考虑如何使用这一工具，而《****》则“计划将”对话人工智能作为其评论的第一道防线。该软件也将是开放源码的，让任何网络开发人员采用它。

这种吹毛求疵的过激行为并不是新现象。例如，用于阻止网络用户访问非法或不适当内容的网站封锁名单经常因无意中封锁性健康网站等关键信息而受到批评。加入机器学习并不能自动解决这些问题，事实上，它可能会通过鼓励毫无根据的信心而加剧这些问题。对话人工智能的实际测试结果必须仔细观察。同时，你可以在这里查看完整的故事。