Transforming unstructured data into insightful stories: The AURA project
Team AURA is a collaboration between journalists and technologists from The Economist, Indian Express, DR (Danish Broadcasting Corporation),& Aftonbladet as part of the 2024 JournalismAI Fellowship. Here, they discuss their progress mid-way through the 2024 JournalismAI Fellowship.
Our project, AURA, efficiently transforms unstructured data into structured insights, leverages advanced language models for contextual understanding and integrates expert knowledge to provide detailed, high-quality reports.
We entered the project with a few different ideas and perspectives but soon settled on a shared vision: to provide journalists with a powerful research tool that would put new research in context, find relevant experts and work as an assistant throughout the process.
Cross-departmental collaboration
One of the most interesting aspects of the fellowship has been the opportunity to work with colleagues from both editorial and technical backgrounds. This diversity has been both a challenge and a strength, as we navigate how best to integrate our skills and perspectives to create innovative solutions. Collaboration can be challenging - especially when collaboration is the starting point, with the end product only determined afterwards. However, in this case, the real value has been the collaboration itself and the chance to connect with colleagues from around the globe. Despite our different backgrounds and realities, we all face similar challenges, and this shared experience has enriched the entire process.
Clarote & AI4Media / Better Images of AI / AI Mural / CC-BY 4.0
Equally important has been the broader community aspect of the Fellowship. Observing the various ways in which AI is being utilised in newsrooms, and the willingness and speed of other organisations to experiment with AI has been incredibly valuable. The exchange of knowledge and insights has been one of the most rewarding aspects of this experience.
Our team includes those with strong editorial expertise, who are skilled in storytelling and audience engagement, and those with technical expertise, focusing on AI development. Additionally, some team members bridge both worlds, providing valuable insights into how technology can support journalistic practices. Our group is truly diverse in every aspect, and the power it creates is inspiring. We come from different backgrounds and have different skills, but agree on a common idea that will hopefully result in something truly useful.
Problem-solving: a sum of the parts
Working across such diverse roles has highlighted different approaches to problem-solving. For instance, editorial team members often emphasise the importance of narrative and audience impact, focusing on storytelling and human elements. In contrast, the technical team members prioritise data-driven insights and technological efficiency, seeking innovative ways to enhance storytelling through AI. The golden learning here is to figure out how to employ all the considerations and still deliver a product within a somewhat tight timeframe. Developing the skills necessary to apply AI within the context of this Fellowship has required us to strike a balance between leveraging the technological advancements available, while also recognising the limitations of the technology to ensure that core principles of journalistic standards are maintained.
This has sometimes led to challenges, such as deciding whether to act first and iterate based on feedback or to engage in thorough discussions before implementing solutions. Balancing these approaches has required open communication and mutual respect, allowing us to find common ground and develop strategies that incorporate the strengths of both perspectives.
Despite these challenges, the interplay between editorial and technical roles has significantly enhanced our project. The first half of the fellowship has taught us the importance of collaboration and flexibility. We have developed strategies to bridge the gap between our different roles, such as setting clear goals, fostering open communication, and creating an inclusive environment where all voices are heard.
The 2024 JournalismAI Fellowship brought together 40 journalists and technologists to collaboratively work on using AI to improve journalism, its systems and processes. Read more about it here.