AI in Indian Newsrooms: Balancing Automation and Editorial Trust

The integration of Artificial Intelligence (AI) into Indian newsrooms is rapidly evolving, presenting both opportunities and challenges for the long-term sustainability of the industry. A recent forum hosted by the World Association of News Publishers (WAN-IFRA) in Bengaluru highlighted a spectrum of approaches, from cautious experimentation to more aggressive adoption, all while grappling with the crucial need to maintain public trust.

The ‘AI Elephant’ in the Room

Prof. Sunil Saxena, Founder-Director of the AI Media Academy, aptly described AI as the “elephant in the room” for Indian media leaders, emphasizing its growing influence on newsroom practices and business models [1]. The urgency surrounding AI adoption was a central theme of the WAN-IFRA AI in Media Forum held February 23, 2026.

Varied Approaches to AI Integration

Publishers are adopting AI at different paces and for different purposes. The Printers Mysore, publisher of Deccan Herald and Prajavani, has primarily focused on utilizing AI for Search Engine Optimization (SEO), data tagging, and coding within its digital and tech teams [1]. While editorial teams have shown some “resistance and curiosity,” the company plans to test AI-powered translation tools to expand content reach.

A “human sandwich” model is being implemented, combining AI assistance with journalist input at the beginning and complete of the journalistic process, emphasizing the importance of human oversight [1].

Collective Newsroom, the BBC’s Indian-language content provider, takes a more limited approach, explicitly avoiding AI for content generation due to the paramount importance of trust associated with the BBC News brand [1]. However, AI is utilized for curation, translation, and basic clip editing, always with clear disclaimers regarding its utilize. The company has as well experimented with using AI to protect the identities of journalists working in countries with authoritarian regimes by transforming their voices.

Thomson Reuters, in contrast, is pursuing a more “aggressive” integration of AI to “augment” its existing capabilities, incorporating it into its content management system (CMS) for proofreading and multimedia asset packaging [1].

Manorama Online aims for an AI integration with a “human touch,” ensuring human supervision at every stage of production and oversight before publication [1].

Navigating the Multilingual Landscape

India’s diverse linguistic landscape presents unique challenges for AI implementation. While AI performance in English and Hindi is improving, a noticeable gap remains compared to its effectiveness in other Indian languages, largely due to a lack of training data for smaller languages [1]. Regional language press is considered more attuned to local nuances and complexities, such as those related to caste, religion, and gender, than content used to train most language models.

Interestingly, publishers like Manorama Online have not experienced a decline in search traffic, likely due to the fact that most of their content is in Malayalam, a language less affected by AI answer engines. However, their English-language title, Deccan Herald, has seen a decline in search traffic, potentially due to AI overviews and chatbots providing direct answers [1].

Building and Maintaining Trust

A key question raised at the forum was whether AI will ultimately increase or decrease trust in media. Panellists generally expressed optimism, viewing AI as a potential ally in building trust if used to assist journalists throughout the newsroom process [1]. AI tools can, for example, efficiently analyze complex legal documents.

The consensus was that newsroom leadership must strategically decide how to use AI, focusing on enhancing journalistic capabilities and safeguarding the interests of journalists, rather than simply cutting costs. The need to “tame this beast” and prioritize core journalistic values was also emphasized, recognizing that AI will likely become even more powerful in the future [1].

WAN-IFRA hosted a hands-on AI workshop in Bengaluru on February 23, 2026, led by Prof. Sunil Saxena, focusing on workflow automation, prompt engineering, and practical newsroom applications [2]. The workshop aimed to support journalists integrate AI tools into their daily workflows using ChatGPT to build newsroom apps without coding [4].