AI and its impact on the future of media

Some claim that AI will make its imprint on the world as the engine powering the Fourth Industrial Revolution. However, there are concerns about harms produced by AI such as discrimination and misinformation, or bias in training sets. As AI-generated content proliferates, cultural critics fear a “doom loop” of AI systems learning from AI-produced content and going rogue or producing outputs of ever lower quality. Others worry about corporate power, and how locating the majority of AI infrastructure in the US and China has produced a gap between how well models perform in English and Chinese compared with other languages. Despite major technical advances, AI productivity gains are not guaranteed. Translating innovation into business cases and making them financially viable is complex.

The relationship between AI and media has been transformative and tumultuous. The rapid development of AI technologies such as GPT-4 has led to sensational stories and hyperbolic claims. Headlines oscillated between extreme optimism for a vibrant future for creative innovation and dire warnings about the death of journalism.

Despite these dramatic narratives, analysis of the practical impact of AI on media has been more nuanced. Instead of rampant job losses, studies from MIT suggest only 0.4% of tasks can be fully automated cost-effectively, implying that AI will complement rather than replace human jobs. This is particularly relevant for journalism, where AI can assist in tasks such as generating headlines or translating content, but human oversight remains crucial to ensure accuracy and ethical standards.
 

AI has quietly revolutionised the advertising industry. According to GroupM, (1) around 70% of ad revenue is influenced by AI, albeit in an “informed” manner that includes statistics and machine learning rather than just generative AI. However, direct AI revenues in advertising are still in their infancy. For example, Google’s AI-driven ad features are rapidly growing but remain a small fraction of its overall revenue. From media planning to optimisation and creative asset generation, AI is bound to affect the entire value of advertising. This is likely to reconfigure the industry, spark new companies and service models and new types of jobs where human expert and technological systems interact naturally.

Source: GroupM June 2024 This Year Next Year
 

Bertelsmann, as a global media conglomerate, offers a practical example of how AI is being integrated into various business domains. Since 2019, the company has focused on leveraging AI for data management, cloud partnerships, and innovative applications. A recent study highlighted that AI-generated video content could constitute up to 50% of all video content by 2027, a startling statistic that underscores the transformative potential of AI.

Bertelsmann has adopted a grassroots approach to AI, encouraging individual business units to explore AI use cases. This has led to numerous projects from which four priority areas were selected: books, music, newsroom and entertainment. In the publishing branch, Penguin Random House, AI is used to identify books in social media feeds and create nearly real time sponsored posts featuring the relevant book in users threads.

These AI-matched campaigns doubled purchase rate and shortened the campaign creation time. At music publisher BMG, AI supports video editors in their use of music by creating prompting tools able to match the tone of a video with a music suggestion, making better use of the BMG catalogue. AI is also a powerful tool for entertainment producers. By applying AI-generated background to RTL shows or enhancing the sound of Fremantle productions, it creates efficiencies for content creators by saving post production time and money. In the newsroom, ntv is testing an AI text generator, able to convert press clippings from news agencies into proper text for the ntv website using the right tone.

Bertelsmann has prioritised responsible AI use across its business units. The company has established an AI Council to oversee ethical considerations and ensure that AI technologies are used responsibly. This aligns with broader industry trends, where AI ethics conferences have seen a surge in participation and academic contributions.

As we move forward, it is essential to embrace the imperfections inherent in AI technologies. Early AI pioneer, Alan Turing, famously stated that “if a machine is expected to be infallible, it cannot also be intelligent.” This sentiment is particularly relevant today, as it emphasises the importance of learning from mistakes to advance AI capabilities.

For aspiring journalists and AI researchers it is crucial to actively engage with AI technologies. Theoretical knowledge must be complemented by practical experience to unlock AI’s full potential. As the media landscape continues to evolve, new forms of storytelling and content delivery will emerge, driven by the relationship between AI and human creativity.
 

By embracing the imperfections and responsibly harnessing the power of AI, we can look forward to a future where AI augments human capabilities, rather than replacing them. For media companies, responsible AI is essential, as is tapping into its opportunities. The future promises a blend of human touch and AI capabilities, ensuring a balanced and ethical approach to technological advancements.