Abstract

With the increase in online content circulation new challenges have arisen: the dissemination of defamatory content, non-consensual intimate images, hate speech, fake news, the increase of copyright violations, among others. Due to the huge amount of work required in moderating content, internet platforms are developing artificial intelligence to automate decision-making content removal. This article discusses the reported performance of current content moderation technologies from a legal perspective, addressing the following question: what risks do these technologies pose to freedom of expression, access to information and diversity in the digital environment? The legal analysis developed by the article focuses on international human rights law standards. Despite recent improvements, content moderation technologies still fail to understand context, thereby posing risks to users’ free speech, access to information and equality. Consequently, it is concluded, these technologies should not be the sole basis for reaching decisions that directly affect user expression.

中文翻译：

---

内容审核技术：应用人权标准保护言论自由

摘要

随着在线内容发行量的增加，出现了新的挑战：诽谤性内容的传播，未经同意的私密图片，仇恨言论，虚假新闻，侵犯版权的行为增加等等。由于审核内容所需的大量工作，Internet平台正在开发人工智能以自动执行决策内容删除。本文从法律角度讨论了当前内容审核技术的报告性能，解决了以下问题：这些技术对数字环境中的表达自由，信息获取和多样性构成什么风险？本文所进行的法律分析着重于国际人权法标准。尽管最近有所改进，内容审核技术仍然无法理解上下文，从而给用户的言论自由，信息获取和平等带来风险。因此，可以得出结论，这些技术不应成为做出直接影响用户表达的决策的唯一基础。