Liam Andrew, Chief Product Officer at The Texas Tribune, discusses building a data warehouse and implementing tools like automatic content tagging and content recommendations using AI. The podcast explores the importance of data in AI innovation and the opportunities it presents for nonprofit newsrooms. It also delves into entity recognition, AI's benefits for journalists in the newsroom, and the impact of AI on journalism and funding opportunities.
Read more
AI Summary
AI Chapters
Episode notes
auto_awesome
Podcast summary created with Snipd AI
Quick takeaways
The Texas Tribune prioritizes building a data warehouse and analytics hub to provide transparency and insights about their work and audience.
The Texas Tribune utilizes AI tools like automatic content tagging and content recommendations to improve content organization, analysis, and reader engagement.
Deep dives
Building a Data Infrastructure at the Texas Tribune
The Texas Tribune prioritizes building a solid data infrastructure for their newsroom. They recognized the importance of good quality data for AI development and implemented an analytics hub to house their data. The hub provides democratic access to all Tribune staff, allowing them to analyze data from various sources such as podcasts, newsletters, and website readers. By building a data warehouse and creating a shared resource for analytics, the Tribune has been able to gain insights about their own work and audience.
Implementing AI Tools for Content Tagging
The Texas Tribune utilizes AI tools, specifically for their automatic content tagging system. They recognized the significance of categorizing their content for meaningful insights and patterns. By using named entity recognition, the Tribune suggests relevant tags to their editors, connecting actors mentioned in stories to specific topics. This helps ensure consistent and accurate tagging across different desks, enabling better content organization and analysis.
Content Recommendation System and Collaboration
The Texas Tribune has implemented a content recommendation system to suggest related articles to readers. Initially, editors manually curated these recommendations, but they partnered with the local news lab at the Brown Institute to develop an AI-driven recommendation system. By using collaborative filtering, the system analyzes what other users have read alongside the current article to provide relevant recommendations. While there was a slight drop in recirculation, the time saved by automating the process outweighed the decrease. The Tribune continues to explore AI-powered recirculation methods to maintain engagement.
AI's Impact on Journalism and Nonprofit Newsrooms
AI is set to have a positive impact on journalism, particularly in nonprofit newsrooms. Many funding organizations and tech companies are expressing interest in supporting AI initiatives within local newsrooms. The future holds potential for newsrooms to build their own language models suited to their specialized coverage, in addition to utilizing existing AI tools. Collaboration and resource sharing within the industry can help facilitate the adoption and customization of AI technology in nonprofit news organizations.
Liam Andrew, the Chief Product Officer at The Texas Tribune joins Nikita Roy to discuss how they built a data warehouse and analytics hub to provide transparency around key metrics for their newsroom. Liam also shares how The Tribune has implemented tools like automatic content tagging and content recommendations using AI.
In his role at the Tribune, Liam leads the newsroom's software strategy and operations, and oversees the engineering, design, and analytics teams. He joined the Tribune in 2015 as a developer, bringing a background in software engineering, product strategy, and user experience design from media startups and academic research labs.
Tune in to hear how a non profit newsroom in the United States has been leading in AI adoption.