ActiveFence + Daily: Helping developers deliver AI-powered content moderation in real-time video

Daily’s developer platform powers audio and video experiences for millions of people all over the world. Our customers are developers who use our APIs and client SDKs to build audio and video features into applications and websites.

Our AI Week series looks at how developers can combine real-time video with AI as they build with our platform. Today we're publishing two more posts in this AI series — first, our daily-python SDK announcement here; and below we take a look at a key use case for AI, content moderation.

You can learn more about the topics we’re covering in this AI Week series, and how we think about the potential of combining WebRTC, video and audio, and AI, in our kickoff post. Feel free to click over and read that intro before (or after) reading this post.

Today, as part of our AI Week series, we're looking at how combining Daily and ActiveFence gives developers flexible, powerful support for content moderation. This developer-first approach combines two industry-leading capabilities:  ActiveFence's robust ActiveScore content moderation APIs alongside our daily-python SDK, a toolkit for creating AI-powered real-time features and applications.

Using ActiveFence together with Daily, product and engineering teams can create seamless solutions to identify unwanted, destructive, and illegal behavior, safeguard users, and meet industry and compliance standards. 

Content Moderation with AI-powered detection

User safety and trust has become a central topic of discussion for online platforms and social applications. Solutions are needed to address issues including bullying and harassment, hate speech, violence, adult content, child exploitation, and illegal goods. Regulatory guidelines also are expanding to help protect users. As of February 2024, the EU Digital Services Act (DSA) will require online platforms with customers in the EU to implement content moderation, removal, and reporting.

ActiveFence's ActiveScore offering gives developers AI-powered tools to identify 16 different kinds of unwanted content, in real time. Using a proprietary database and tuned AI models, these APIs can analyze text, images, audio and video to provide risk scores and indicate unwanted activity. 

The daily-python SDK gives developers the tools to leverage ActiveFence in their video and audio apps. Briefly, daily-python provides developers:  

  1. The ability to run a real-time Python media-processing client on any cloud platform
  2. Daily client APIs that function in the same ways as our JavaScript, Android Kotlin, and iOS Swift SDKs
  3. Helper functions that bridge between a real-time WebRTC context and common AI tools and services

This means developers can moderate content with ergonomic APIs that allow for maximum customizability. Our teammate Chad has built out an example repo for you to play with. He's also put together this fun clip to to showcase how you can seamlessly integrate content moderation capabilities within your product experience — and test the content detection with some harmless examples:

Supporting content moderation for every developer 

One of our north stars at Daily is providing developers different ways to build, with great resources. Engineering teams can customize and optimize for what's best for their users, product, and mission. Combining daily-python with an industry leader like ActiveFence lets developers create the content moderation UX their tools need. 

It's key that a resource like content moderation be available to every platform. Both ActiveFence and Daily provide free minutes for developers to get started and offer competitive pricing at scale. Let us know how we can help support you as you build. Reach out to our developer support or peerConnection, our forum, or find us online or IRL at one of the events we host. And check back this and next week for more in our AI Week series.

Never miss a story

Get the latest direct to your inbox.