#IBC2025
17
IBC2025 ACCELERATORS: A FRAMEWORK FOR GENERATIVE AI; ECOFLOW II
Two Accelerator projects will be presented on the Future Tech Stage in Hall 14 today – ‘A Framework for Generative AI’ and ‘Ecoflow II’. In an era when AI can spin out videos and animations from simple text prompts, broadcasters are asking how to harness this power without losing creative vision. “The keyword is creative control – and the other keyword is visual consistency,” says Roberto Iacoviello, Lead Research Engineer at RAI and Co-leader of ‘A Framework for Generative AI’. The project takes the lessons of last year’s ‘Generative AI in Action’ initiative and turns them into actionable design goals. “What we learned is that creative people want to maintain control,” says Iacoviello. He emphasises that the ability to direct, refine and define visual elements is not negotiable. “That means shot- to-shot consistency – the characters, movement, environment and so on.” This year’s project aims to build a unified generative AI platform for broadcasting that will let producers input data, prompts and assets to rapidly generate scripts, highlights, short-form promos, TV ads and other content – making the production pipeline faster and more flexible while keeping humans firmly in the loop. The project is supported by Champions VRT, RAI, YLE, Globo, EBU, Channel 4, University of Kent and ITV, and Participants Animatix, Google, AMD/HP, Studio Deussen, Pluxbox and Yaaramarchiano.
The framework covers a range of use cases that focus on every stage of production, from pre-production and scripting through to editing, and even addresses dynamic marketing needs. The team wants to create an environment
where creative professionals – not technologists – drive the storytelling. “It’s not about asking for answers from the tool,” Iacoviello says, “but setting a direction and listening for the responses. It’s a co-creation tool.” Another major goal is the orchestration of workflows. The framework aims to map and support complex, multi-step creative pipelines – from script to storyboard, from video generation to voice-over and music – in a modular, extensible fashion. This approach not only boosts creative agility but also opens the door to highly flexible team collaboration. At IBC2025, a demo app is allowing attendees to generate content using standardised prompts and compare outputs across platforms.
However the highlight of the showcase is the premiere of ‘Echoes of Rome: The Code of
Roberto Iacoviello, RAI
Empathy’, a photorealistic, AI-assisted trailer that demonstrates how the framework handles character consistency, emotion, pacing and scene progression over a sustained narrative. Jouni Frilander, Innovation Lead at YLE, says that this showcase isn’t just a tech demo. “We want to tell a story that resonates, that has cultural weight. It’s not about tricking people into thinking it’s human-made. It’s about what humans can do when empowered by these tools.”
“We want to tell a story that
resonates, that has cultural weight. It’s not about tricking people into thinking it’s human-made. It’s about what humans can do when empowered by these tools” Jouni Frilander, YLE
This afternoon sees the results of ‘Ecoflow
II’ presented. Building on the exploratory work of last year’s Ecoflow I project, the next phase supports efforts across the media industry to better understand and reduce the environmental impact of digital content distribution. This is a highly collaborative project with participation from across the industry.
Champions involved are the BBC, BT Media & Broadcast, Channel 4, EBU, IET, ITV, Bouygues Telecom and Greening of Streaming, alongside Participants Accedo, Bitmovin, Humans Not Robots and Quanteec.
‘Ecoflow II’ aims to establish standardised methods for measuring streaming’s energy consumption, and identify energy-saving opportunities throughout the technology supply chain, continuing the development of data-driven methods, tools and insights to help inform more sustainable technology decisions. The project focuses on the IP distribution network, from origin server to receiving device. The goal is to create a transparent data framework to measure energy use across the media distribution chain, alongside developing scenario-based models to explore how different configurations impact energy use. It also aims to test real-world technologies to reduce energy consumption, gather data on their effectiveness, and share findings with the industry through public demos and forums.
A Framework for Generative AI takes place from 12:30-13:30 on the Future Tech Stage in Hall 14. It is followed by Ecoflow II from 16:30-17:30.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88
