How we report, correct, label, and respond. Plain-language editorial guidelines, machine-checked at publish time.
We try to get things right. When we don't, we say so, on the same page where readers found the original story. We label every form of AI assistance, every conflict of interest, and every anonymous source — not because regulators ask us to, but because reader trust depends on it.
This page is the canonical statement of those standards. Every disclosure on every NTCMS article links back here. The vocabulary is machine-checked: a story can't be published if a declared conflict has no disclosure, or an accusation has no record of contacting the subject, or AI synthesis has no reader-facing notice.
Every published correction is logged on a public page: /corrections. Reader-submitted complaints that result in a correction are linked there with the editor's response.
India's IT Rules 2025 amendment requires synthetic-content disclosure. We follow that as a floor, not a ceiling.
Every article that used AI assistance — for translation, summarization, image generation, voice synthesis, headline drafting, or copy editing — carries a visible disclosure on the article page. We label 7 distinct categories of AI involvement, including the category, what it did, and a "read our AI policy" link.
Full policy: /about/ai
If we report something against a named person or organisation, we contact them and record what they said — or that they declined to comment, or didn't respond. Per BBC Editorial Guideline #4 + IPSO Editors' Code Clause 2.
The publish gate blocks any accusatory story until the right-of-reply status is one of 6 resolved states (with subject contacted, contact timestamp, and — when relevant — verbatim response or explicit waiver).
The reader sees a transparency block on the article telling them which state applied.
If the author or editor of a story has a financial stake, family relationship, paid engagement, prior employment, or advocacy role connected to the subject, we disclose it on the article. 8 categories of conflict are tracked. High-severity conflicts (financial, familial, paid engagement) require reader-facing copy explaining how we managed the conflict.
Full policy: /about/conflicts-of-interest
We use anonymous sources only when identifying them would put them at risk — physical safety, legal jeopardy, retaliation, or as protected whistleblowers. 7 reason categories are tracked.
Per BBC Editorial Guideline #6: at least one editor knows the identity of every anonymous source. Source identities are encrypted at rest with AES-256-GCM, decryptable only by the original reporter, the verified editor-on-record, the editor-in-chief, and admins. Every decryption is logged with the editor's reason and retained for ten years.
We do not respond to legal demands for anonymous source identity without first notifying the source where doing so is lawful. Bulk export and disclosure-status changes require dual approval from the editor-in-chief and admin.
If you spot an error, a privacy concern, a story you think was unfair, or content that should be taken down: /editorial/complaint.
Every accepted complaint gets a ticket id, an SLA, an editor, and a published response — even if the response is "we got it right and here's why." We track 8 complaint categories with kind-specific SLAs aligned with India's IT Rules 2021 (15-day grievance redressal) and DPDP Act 2023 (30-day data export/delete).
Status of any submitted complaint: /editorial/complaint/status
This charter aligns with:
Vocabulary is maintained in config/ai_disclosure_kinds.php, config/coi_disclosure_kinds.php, config/right_of_reply_policy.php, config/editorial_complaint_policy.php, config/anonymous_source_ledger_policy.php. Changes are governed via the config/sensitive_action_registry.php dual-approval system.
Last reviewed: May 2026.
ऐसा लगता है कि आप ऐड ब्लॉकर इस्तेमाल कर रहे हैं।
कृपया ऐड ब्लॉकर बंद करके हमारा सहयोग करें।
इस साइट के लिए ऐड ब्लॉकर बंद करने के लिए ये कदम अपनाएं:
पूरा होने के बाद नीचे क्लिक करें।