Artificial intelligence may be transforming newsrooms, but its most valuable role is not writing articles. According to Srinivasan Ramani, deputy national editor at The Hindu, its real power lies in helping journalists process and structure data at a scale no reporting team could manage alone.
Speaking at WAN-IFRA’s 2026 AI in Media Forum, Ramani described the technology as a “sophisticated intern” , a tool that can sift millions of records, generate queries and accelerate development work while reporters retain editorial judgement.
For publishers grappling with how to deploy AI responsibly, the approach offers a pragmatic model: use machines to handle scale and structure while journalists define the questions and the story.
The scale of the newspaper’s recent investigation illustrates the shift. Reporters analysed nearly 22 million voter records across Bihar, Tamil Nadu and West Bengal while examining India’s Special Intensive Revision of electoral rolls. The data was difficult to use , records existed only as image-based PDFs in Hindi.
Using optical character recognition and large language models to generate SQL queries from natural language prompts, the team converted about 250,000 files into a searchable database. The analysis uncovered anomalies, including unusually high deletion rates among voters under 50 who were marked as deceased.
“The hypothesis was ours,” said Ramani. “AI helped us process the scale.”
The findings were raised in Parliament and later referenced in court proceedings.
The same approach has reshaped how the newsroom builds digital interactives. During India’s 2024 general election, the team created a results interface with constituency maps and rural and urban filters using tools including ChatGPT, Gemini and Claude. The prompts generated much of the code needed for the project, reducing what would typically require a team of developers to a two-week sprint.
Experimentation has also extended beyond software. While reporting on extreme heat in Chennai, journalists built low-cost environmental sensors using Arduino hardware. The devices , assembled with guidance from AI tools , recorded heat exposure experienced by fishermen and rickshaw drivers every ten seconds.
For Ramani, the lesson is simple: AI works best when it augments reporting rather than replacing it. Used as an analytical assistant, it allows journalists to tackle investigations and data projects that would otherwise be impossible for most newsrooms.
Supercharge Your Content Strategy
Feel free to test this content on your social media sites to see whether it works for your community.
