The 2025 edition of the International Broadcasting Convention brought over 45,000 professionals to Amsterdam in September, with a clear message: the experimentation phase with artificial intelligence is over. Now comes the hard part.
Walking through the 14 exhibition halls at RAI Amsterdam, you could feel a shift in the conversation. Gone were the breathless predictions about AI transforming everything overnight. Instead, leaders were asking more challenging questions about production, industrialisation, and actual return on investment.
This evolution reflects a maturing industry that has learned some expensive lessons about the gap between pilot projects and production systems. The path from proof of concept to real business value remains littered with abandoned initiatives that looked impressive in demonstrations but failed to scale and deliver.
From panic to pragmatism
The pattern has become familiar across sectors. In 2024, boards demanded AI strategies. C-suites, caught unprepared, scrambled to demonstrate activity. Teams frantically developed use cases to show they were "doing something with AI" without first establishing clear links to business value or production pathways.
The result was significant time and resources expended on projects that never delivered meaningful returns—the recent MIT research, showing that 95% of generative AI projects fail to deliver ROI (albeit using a sample size of only 300 organisations), underscores this point. The technology often worked perfectly well in controlled environments. The business case simply wasn't there.
This year's conversations at IBC revealed a welcome shift. Organisations have moved from asking "what can AI do?" to "where does AI solve genuine business problems?" The distinction matters enormously.
Some organisations are starting to get this right. One major UK broadcaster shared at Cognizant's roundtable with Google Cloud how they used AI to democratise advertising access for smaller businesses. What would typically cost around $300,000 became available for approximately $10,000. The small business received effective advertising. The broadcaster's commercial teams saw new market possibilities.
This crystallises the new approach: identifying specific business challenges where AI can deliver measurable value. Technology serves the business need rather than the other way around.
Added value through AI
Another example emerged from Spanish-language broadcasting, where legacy content archives represent untapped value. AI-powered metadata categorisation transforms decades of licensed content into searchable, reusable assets. This creates potential revenue streams from existing libraries that would otherwise remain dormant.
The challenge extends beyond identifying opportunities. Productivity savings often prove difficult to quantify unless organisations actually reduce legacy resources and operating costs. Simply automating existing processes without optimising the broader workflow delivers limited benefit.
Revenue generation through AI remains elusive for most organisations. A streaming executive at IBC described broadcasting a major sporting event this summer using AI to generate personalised highlights. Customers loved the feature. They expected it for free.
This captures a fundamental tension in the deployment of AI. The technology makes creating sophisticated features easier and cheaper. It simultaneously makes those features harder to monetise as they quickly become table stakes rather than premium offerings.
Customer expectations move faster than business models can adapt. Today's innovative feature becomes tomorrow's baseline requirement. Organisations must factor this acceleration into their AI investment calculations rather than assuming sustained differentiation.
The two-sided nature of AI economics demands careful consideration. Cost savings can be substantial and relatively predictable. Revenue upside proves far more uncertain as competitive dynamics compress premium features into standard offerings.
Trust and transparency matter
Trust emerged as a recurring theme across IBC discussions. Audiences quickly detect and reject content that feels artificial or manipulated. News teams, in particular, struggle with AI integration, recognising that credibility damage from perceived "fakeness" can be catastrophic.
This represents more than reputational risk. It touches fundamental questions about content authenticity and audience relationships that media organisations have spent decades building. AI deployment must strengthen rather than undermine these foundations.
Transparency provides the obvious starting point. Leading organisations are clearly flagging AI-generated content, publishing educational materials about deep fakes, and investing in audience education. Some even advocate for regulated AI categorisation systems similar to technical standards like UHD displays.
The analogy resonates. When consumers purchase devices with specific technical certifications, they expect to meet certain performance standards. AI categorisation could provide similar clarity about what AI actually does within specific applications and whether it delivers genuine value beyond marketing claims.
This regulatory standardisation would benefit both consumers and producers. It would help organisations differentiate genuine AI capabilities from superficial labelling, while providing audiences with clear frameworks for evaluating authenticity and quality.
Partnership and execution
The complexity of modern AI deployment makes partnership essential. Even large organisations around the IBC roundtable acknowledged their reliance on external expertise. The technology evolves too quickly for any single organisation to maintain comprehensive internal capabilities across all areas.
The strategic question becomes where to focus internal resources versus where to leverage external partnerships. Organisations need to identify the specific parts of their workflows that provide genuine competitive differentiation and keep those in-house. Everything else becomes a candidate for partnership.
This approach requires an honest assessment of organisational capabilities and strategic priorities. Many organisations struggle with this discipline, attempting to maintain control over too many elements while lacking the resources to execute them effectively.
Finding the right partners is crucial, given the rapid pace of technological change. Organisations need partners who understand media industry dynamics, possess deep technical expertise, and can scale solutions from pilot to production. The quality of the partnership often determines success more than the initial technology choice.
The next six months will separate organisations that genuinely adapt from those that talk about transformation. The technology will continue evolving in ways we cannot fully predict. Success will come to those who maintain open minds while grounding decisions in clear business logic.
This isn't hype. The changes are real and accelerating. However, winning strategies will be built on fundamentals: understanding customer needs, delivering measurable value, building trust, and partnering effectively.
Organisations within the media industry that move decisively while maintaining strategic discipline will establish positions that become increasingly difficult to challenge. Those who treat AI as a compliance exercise or continue chasing pilots without production pathways will spend years catching up.
The technology is ready. New business models are emerging. The question now is which organisations will execute effectively on the opportunity that everyone can finally see clearly.
If you would like to learn more, please contact us.
