Last month I put a question to an AI assistant that I would normally ask a well-networked journalist: which organisations carry genuine authority on green buildings in India? The answer arrived in seconds, fluent and self-assured. What held my attention was the list of sources underneath it. Business dailies, trade reports, a few expert bylines. Hardly a brand website in sight.
After a decade of pitching editors across India and the Gulf, I have learnt to be suspicious of anything that flatters my profession. This one earned a second look anyway.
Generative engines assemble answers from material they consider trustworthy, and trust is a scarce commodity online. Anyone can publish. Very few can get an independent editor to put a name behind a claim. That filter, the one PR professionals spend careers learning to pass, turns out to be exactly what these systems lean on. Journalism, expert commentary, contributed articles and third-party mentions carry weight because a human with reputational skin in the game decided they deserved space.
Think about what this does to the old hierarchy of channels. For years many of us defended earned media with a slightly apologetic shrug, since coverage is slow to land and hard to tie to revenue. Paid and owned channels, meanwhile, came with dashboards. Now a machine is quietly voting for the slow channel.
The Indian context sharpens the point. Our publishing ecosystem is dense: national dailies, sector journals, specialist portals, regional language outlets. Much of it stays online for years. A feature placed in a respected outlet in 2024 may still be shaping how an AI tool describes your company today, long after the campaign report was filed and forgotten. The inverse is equally true. Where credible coverage is thin, the engine fills the vacuum with whatever it finds, and that could be a rival's framing, an outdated profile or an anonymous forum thread.
Sectors with long buying cycles feel this first. Trade, financial services, healthcare, sustainability and enterprise technology all involve buyers who research for months before speaking to anyone. More of that research now begins in a chat window. A company with a deep trail of independent coverage gets described accurately. A company without one gets described by whoever wrote more.
For communications teams, I see a few implications, and none of them needs a new budget line.
First, the homepage has moved. Buyers, investors, candidates and reporters increasingly ask an AI tool for a summary before visiting any website. If that summary is the first impression, the coverage feeding it is the real front door. Most teams I know track web traffic obsessively and have never once checked what an AI says about their own brand.
Second, consistency has acquired a measurable payoff. These systems read for patterns. If a spokesperson is a water technology authority in one article, a supply chain commentator in another and a general business voice in a third, the signal blurs. Message discipline used to be an internal hygiene matter. It now has a visible external consequence.
Third, placement quality outranks volume. I say this as someone who has sat through many meetings where coverage counts were waved around as proof of success. Twelve syndicated releases on low-authority sites add little. One byline with a real argument, in a publication editors respect, adds a great deal. I have always preferred fewer and better placements, and now there is a sharper argument to take into the boardroom.
It also changes how I answer founders who ask whether a byline is worth the effort when LinkedIn offers instant reach. A post travels for a day. An article on a credible outlet's site can keep informing search and AI answers for years. Both earn their place. They just do different jobs.
If you want a place to start, begin with an audit. Ask the major AI tools about your company, your category and your senior spokespeople. Record which sources they cite, what they get wrong and what they omit. Then check that leadership bios, company facts and expert positioning agree everywhere they appear, owned and earned alike. Finally, add AI visibility to your reporting next to share of voice and message pull-through. Even a rough quarterly check will teach you more than you expect.
A caveat is due. Nobody outside the model builders knows precisely how these systems weigh their sources, and the rules shift often. Anyone promising a guaranteed formula deserves a raised eyebrow. What can be said with confidence is directional: credible, independent and well-structured content gets noticed.
The conversation about AI writing our content has its place, and I follow it closely. Beside it sits a quieter question that I think matters more to our profession: when machines read the world's content, whom do they choose to believe? Earning belief from journalists, analysts and audiences has long been our craft. The craft has stayed the same. The audience just gained a new member, one that never skims.
(Zainab Chakkiwala is a Senior Manager at AZN Partners, a PR and communications agency, with over a decade of experience across India and the Middle East.)
