At this year’s NAB Show, the almost child-like excitement around AI revealed a collective belief in its transformative potential for our industry. The AMPLIFY bar in the West Hall, for example, was abuzz with discussions on AI's promise to change how content is created, delivered, and consumed.
A May 2024 McKinsey report showed that 65% of the organizations it surveyed actively use generative AI, and this number has doubled over the previous 12 months. The question for media organizations is not about whether they should use AI - but how they should do so effectively.
As the media landscape evolves, content owners, broadcasters, and streamers must adapt to these changes to stay relevant and competitive. Understanding and implementing AI is no longer optional—it's essential. Treating AI as an integral component of operations can enhance content value, increase monetization, and improve operational efficiency. It offers powerful solutions to pressing business challenges, opening new avenues for growth and enabling a competitive edge.
Viewers Expect Personalized Experiences; Creatives Require FreedomThe shift towards streaming and video-on-demand has rendered traditional content delivery and monetization methods increasingly inadequate. AI provides the tools necessary to navigate this complex media environment, addressing critical issues and enhancing viewer engagement.
According to data from Google Cloud, 81% of viewers expect streaming services to provide highly personalized experiences. Making content more engaging and stickier helps build viewer loyalty, ensuring that audiences return time and again. Tailoring content to individual preferences is not a luxury; it’s necessary to retain viewers and stay competitive.
Today, AI can enhance the value of content, expand metadata coverage in archives, and much more."
AI’s ability to analyze viewer data and offer customized recommendations is key to meeting these expectations and reducing viewer churn. Tailoring content to individual preferences is not a luxury; it’s necessary to retain viewers and stay competitive. AI’s ability to analyze viewer data and offer customized recommendations is key to meeting these expectations and reducing viewer churn.
Today, AI can enhance the value of content, expand metadata coverage in archives, and much more. AI can also reduce manual operational costs, but this does not fully alleviate the wider pressures facing media businesses, which includes fully optimizing their workflows.
Traditional processes like video editing, highlights packages, metadata generation, and quality control are often time-consuming and labor-intensive. AI can streamline these operations, allowing more resources to be allocated to strategic initiatives that drive revenue growth and innovation—often by using a hybrid approach with AI providing a first pass, with human creativity used to verify and be creative. Automating repetitive tasks with AI frees creative talent to focus on what they do best - enhancing the overall production process.
AI Has Been Around for a While...What’s Different Now?
Traditional AI in the media technology space focused on applying machine learning (ML), which uses training data to enable systems to learn, improving from experience without being explicitly programmed for each task. This enabled better processing enhancements for efficient video encoding, dynamic resource allocation, super-resolution upscaling of video, and fast data analysis to derive predictions that can be fed into recommendations or customer care platforms. While ML has improved efficiency and quality, generative AI technology is set to have an even greater impact on the future of media.
Generative AI, which essentially mimics human intelligence, enables radical innovation and a level of speed and quality in output that ML alone can rarely achieve. For example, AI-driven content translation and localization, with careful tailoring of instructions can produce stunning translated commentary.
Commentary that retains the original speaker’s tone and timbre but with the intonation and grammar of a different language, all while preserving the ambiance and background of the stadium! This capability could allow media companies to connect with diverse audiences by offering content in multiple languages and enhancing accessibility. It ultimately comes down to audience preference: viewers can opt for native language commentary or an AI translation that captures 95% of the original.
Monetizing content effectively is another area where AI shines. AI can optimize ad placement by providing insights into audience behavior, and ensuring that advertisements are contextually relevant and more likely to resonate with viewers. Picture a game where a celebrity is seen drinking a Coke, followed by a perfectly timed Coke ad. This level of personalization can significantly boost ad effectiveness and viewer engagement.
Further Challenges: Hyperscaling and Societal Acceptance
The efficiency gains from AI are substantial. AI-powered tools enhance video encoding and compression, improving quality while reducing bandwidth requirements. Automated workflows expedite tasks like auditing and reporting, providing faster turnaround times and more accurate insights. These efficiencies translate to significant cost savings and improved profitability. According to McKinsey, high performers in generative AI report that these efficiencies allow them to attribute over 10% of their profits to AI deployment.
But there’s a catch—the creation of generative AI models can require the levels of resources and data that only the Hypescalers can access. Leveraging some of the best AI capabilities will almost certainly mean connecting to AI tools in multiple hyperscale cloud environments and working agilely, updating and swapping models and tools to meet changing business goals.
There are also social acceptance issues to consider. What will audiences accept? Would they prefer to know that the content is in some way enhanced by AI or not? Additionally, 44% of organizations have faced negative consequences from generative AI (McKinsey), including inaccuracies and cybersecurity threats. Generative AI, like humans, can make errors, so expectations that software-based processes will always behave predictably and correctly need to be managed.
In today’s world, AI can be a cause of fake content, but equally, it provides the best potential for detecting fakes. This is an area of increasing importance, which is being address by the industry body C2PA, in order to provide a means to authenticate genuine content and validate its provenance.
Empowering Developers
AI accelerates innovation, allowing rapid deployment of new features and enhancements. This agility enables developers to test multiple functionalities simultaneously, gather real-time feedback, and iterate quickly. The ability to experiment rapidly without wasting time and resources ensures developer innovations are aligned with market needs. Moreover, AI-driven insights guide development, reducing the risk of investing in non-resonating features and ensuring efforts are focused on what matters most.
Data ownership and management needs careful consideration to ensure businesses retain control over their valuable data and intellectual property and don’t just hand the insights to third parties. Building multi-layered workflows that have the flexibility to integrate different AI models (including those developed by the customers themselves) into existing processes allows for flexibility and scalability. Empowering both developers and non-developers with accessible AI tools ensures broader team contributions to AI initiatives. Ongoing training and expert support are vital for robust, scalable AI implementations.
The path to AI integration may seem complex, but it can be navigated effectively with the right approach and partners, leading to significant rewards. AI is not just another technology; it’s a change in mindset for the media workflow, offering the power and versatility to address business challenges with radical innovation.
The Importance of Embracing AI
Integrating AI into media workflows is not a futuristic concept but a present-day necessity. Embracing AI allows media companies to meet the demands of today's viewers and stay ahead of tomorrow's trends. The real challenge lies in harnessing AI's potential , as its true power lies in its versatility. With the right tools, AI can solve diverse business problems, from enhancing audience engagement to streamlining operations and driving innovation. AI frameworks can be customized to address specific business challenges, whether it's improving ad targeting, optimizing content distribution, or enhancing viewer analytics.
Data and content created by AI is additional value that can be derived from original content and audiences, it becomes a key asset for whoever owns it.
AI models should be seen as another tool that can be enhanced, changed, or modified to achieve goals, rather than a huge, customized workflow that only serves to feed that model. As a business grows, AI solutions should scale and adapt with them, ensuring that AI investments continue to deliver value.
