Artificial Intelligence in Media Part 2: Calm reflections with Sanjeevan Bala, ITV Chief Data & AI Officer
Nov 17, 2023
auto_awesome
Sanjeevan Bala, Chief Data & AI Officer at ITV, discusses the opportunities and challenges of AI in media, including running immersive AI workshops with ITV leadership, commoditization and consumerization of AI capabilities, and when to partner, build, or buy AI solutions. They also touch on integrating AI into data strategy, the future impact of AI in the media industry, aligning data strategy with business goals, and the importance of having a business strategy for AI initiatives.
Media organizations have a unique advantage in leveraging AI and data by fine-tuning large language models with their own content assets.
ITV aligns data and AI initiatives with their business strategy to ensure they contribute directly to growth and involve colleagues in demystifying AI.
Deep dives
The Role of Large Language Models
Large language models are described as being an inch deep and a mile wide, meaning they have broad knowledge but lack depth. Media organizations, on the other hand, have a narrower but deeper understanding of their content and its cultural significance. The IP and content assets owned by media organizations give them a unique advantage in leveraging AI and data to drive value. While large language models have their uses, the real value comes from fine-tuning and enhancing them with the media organization's own content assets, which contain nuanced language and cultural context.
Approach to AI and Data at ITV
ITV's approach to AI and data is driven by their business strategy. They focus on aligning data and AI initiatives with their key business goals, such as doubling the number of monthly users, viewing hours, and digital revenue. By doing so, they ensure that technology and data efforts directly contribute to the growth and success of the organization. ITV emphasizes the need for a people-centric approach, involving colleagues across the business in understanding the possibilities and demystifying AI. They have also established clear guardrails to ensure responsible usage of AI and data, addressing concerns related to IP ownership, model transparency, and ethical considerations.
Identifying Opportunities and Use Cases
ITV has identified numerous use cases for AI and data across their value chain, including content production, promotion, and monetization. They are exploring opportunities for personalization and high-level personalization in marketing, leveraging data to tailor promotions and reach viewers both on and off their platforms. ITV aims to identify 10x opportunities, looking beyond marginal gains and focusing on significant productivity and innovation gains. They take a business-driven approach, mapping value cases and considering the potential impact and scalability of different solutions. ITV also acknowledges the importance of integrating AI tools into existing workflows to maximize productivity and efficiency.
Building Trust, Deciding to Partner or Build
In deciding whether to partner or build AI solutions, ITV considers their core competencies, agility, time to business value, and scalability. They assess the benefits of partnering with established tools that already integrate into their workflows versus building internal solutions. ITV recognizes the ongoing maintenance and support required for internally built solutions, comparing it to the potential agility and scale achieved through partnerships. Building trust between media companies and vendors is crucial, as AI technologies have implications for reputation, consumer perception, and IP protection. ITV emphasizes the need for responsible and ethical use of AI, aligning AI initiatives with their organizational purpose.
Sanjeevan Bala is ITV Chief Data & AI Officer. In Part 2 of the DPP's calm reflections about artificial intelligence in the media industry, Sanjeevan speaks with DPP Editorial Director Edward Qualtrough to offer the content provider perspective about AI in media. He discusses:
The genus of his Chief Data & AI Officer role [1min 47sec]
Identifying opportunities and developing guardrails [5min 01sec]
Running immersive AI workshops with ITV leadership [8min 54sec]
Commodisation and consumerisation of AI capabilities [12min 50sec]
Balancing cost and value; and comparing 10x vs marginal opportunities [16min 10 sec]
When to partner, build and buy with AI solutions? [20min 36sec]
'Taking the model to the data' and the roadmap to AI at the Edge [23min 55sec]
Why you don’t need an AI strategy [27min 29sec]
Macro trends and conjecture about the future impact of AI in media [29min 55sec]
In Part 1 we spoke with Johanna Björklund, Co-founder and CTO at media technology startups Adlede and Codemill, as well as Associate Professor in Theoretical Computer Science with a focus on semantic analysis of multimodal data at Umeå University. You can listen to that episode here.
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode