Microsoft has recently launched Azure AI Content Safety, a new service designed to help users identify and filter harmful content generated by AI or users across various applications and services. This service includes text and image detection, enabling the identification of offensive or undesirable content such as profanity, adult content, violence, and certain types of speech. The aim is to create a safer digital environment and promote responsible use of AI while safeguarding the well-being of individuals and society as a whole, as stated by Louise Han, the product manager for Azure Anomaly Detector.
Azure AI Content Safety offers comprehensive safety measures by providing users with the ability to handle different content categories, languages, and threats. It applies AI algorithms to scan, analyze, and moderate visual content, ensuring a 360-degree approach to content safety. The service is also equipped to moderate content in multiple languages and uses a severity metric to assess the severity of specific content.
On a scale from 0 to 7, content graded 0-1 is considered safe for all audiences. Content graded 2-3 may contain prejudiced, judgmental, or opinionated views. Medium severity content is graded at 4-5 and includes offensive, insulting, mocking, or intimidating language, while high severity content, graded 6-7, promotes harmful acts or endorses extreme forms of harmful activity towards identity groups.
Azure AI Content Safety incorporates multicategory filtering to identify and categorize harmful content across various critical domains, including hate, violence, self-harm, and sexual content.
According to Han, it is crucial to address more than just human-generated content when it comes to online safety, especially with the rise of AI-generated content. Ensuring the accuracy, reliability, and absence of harmful or inappropriate materials in AI-generated content is essential to protect users from misinformation and potential harm and to uphold ethical standards and build trust in AI technologies.
Azure AI Content Safety is priced on a pay-as-you-go basis. For more information on pricing options, interested users can visit the Azure AI Content Safety pricing page.
Unique Perspective:
The introduction of Azure AI Content Safety by Microsoft marks a significant step towards creating a safer digital environment for users. As the use of AI becomes more prevalent in generating content, addressing the potential harms and risks associated with it becomes crucial. By offering a comprehensive set of safety measures and the ability to filter and moderate various types of harmful content, Microsoft aims to promote responsible use of AI, protect individuals, and build trust in AI technologies. The availability of Azure AI Content Safety is also a reminder of the importance of upholding ethical standards in technology and ensuring the accuracy and reliability of AI-generated content.
