Artificial intelligence has been an area of concern and debate in the media ecosystem, as some fear that AI will replace humans in the creative process.
Despite excitement around early versions of generative AI, the most recent tools focus less on generating content and more on improving workflows and augmenting processes. Examples include editing audio and video, compression, encoding and other post-production services, such as mastering and real-time transcription.
The next iterations of AI will make a significant impact across the media creation and delivery ecosystem. We expect the biggest effects to be felt across content creation, distribution and consumption mechanisms, including recommendation engines and e-commerce platforms. AI will help augment the creative process, augment tedious tasks and predict and analyse data throughout the ecosystem.
Content creation
In the context of media production, AI-driven creation tools help lower the barrier of entry into creative processes. But they’re not a replacement for content creators; it’s best to think of this technology as a multiplier. The most powerful uses for AI are to enhance and improve workflows that allow human creators to enhance their capabilities. Some companies are using AI powered by machine learning for image recognition and video analysis— for example, to automatically tag who is speaking on camera at what time, offloading an otherwise tedious manual task.
Recommendation engines
Content recommendation is another key area, encompassing applications from top-tier over-the-top (OTT) streaming providers recommending studio-produced content to social media networks recommending user-generated content. In both cases, AI recommends content based on user preferences and interests, largely determined by online behaviour.
As the universe of available content expands, it becomes challenging for viewers to find what they want to watch. AI provides a powerful content discovery tool, but this is not a new application. Many major content providers have been using ML tools for years. As AI continues to evolve, it may begin to work across different platforms and services based on comprehensive user behaviours and preferences.
There is a growing need to ensure these systems are unbiased and ethical. That involves developing AI that audits itself for fairness. To further increase trust, AI systems should give users more control over how their data is collected and used. This can include options to monitor and adjust what data is considered in their recommendations.
Distribution mechanisms and compression
Another way AI can leverage recommendation systems is to better manage their content delivery network (CDN) by predicting user demand and then placing content as close as possible to the users who consume it. Placing content closer to customers at strategic server locations optimises streaming performance.
The exploding volume and variety of content—particularly in immersive video—creates a need for advanced compression. Visual data compression helps lower the bandwidth for transmission and storage. AI will help power these advanced compression methods. Since different compression formats exist, it is also important to ensure compatibility and interoperability. Everything from the creation tools to the distribution networks to the end-user devices must support the various decoding and playback capabilities. AI can help manage the complexity associated with this orchestration of elements.
Targeted advertising & e-commerce
Advertising can be highly targeted by using AI algorithms to build behavioural profiles on viewers. These profiles help advertisers deliver ads that are relevant to their interests. This is broadly referred to as “programmatic advertising” today. This technology is evolving. AI is being used to create and customise different versions of advertisements for individual consumers. Highly targeted ads, including clickable videos, are a very effective e-commerce solution.
As AI continues to evolve, new applications will continue to emerge.
