Meta, AI, and the Hindu Narrative: A Meta Insider Speaks Up

I spent nearly seven years working at Meta, and during that time, I gained an insider’s perspective on how content moderation works—and doesn’t work. 

So when Mark Zuckerberg recently announced that they are dismantling a large percentage of its human content moderation teams and replacing centralized fact-checking teams with user-generated community labeling, I was definitely intrigued. 

In true Meta-esque fashion, these changes have been cloaked under the garb of promoting more ‘free speech’. While on the surface this may sound like a step forward for free speech, for us as Hindus, caution is required.   My experiences as a Meta employee, especially starting 2020, revealed troubling trends that disproportionately affected Hindu creators and content related to Sanatana Dharma.

A First-Hand Account of Bias

During my time at Meta, I frequently received messages from creators with pages dedicated to Hinduism, seeking help with their accounts being demonetized or shadow-banned. 

One case that stood out was a creator with a significant following who shared videos on ancient Hindu temple architecture. This creator reached out to me desperately seeking help, frustrated about repeated demonetization and shadow banning. I had seen his videos myself- they showcased the architectural brilliance of ancient Hindu temples, delving into their history and possible scientific relevance. His content was thought-provoking, even leaving room for debate and skepticism. Yet, his page was flagged multiple times as “problematic,” leading to severe restrictions.

I escalated the issue internally, and discovered that his page had been “mass-reported” by thousands of users alleging he spread “misinformation”. The algorithm flagged him as a repeat offender, making his content vulnerable to future penalties with even minimal reporting. This wasn’t just a technical glitch; it was the result of deliberate manipulation by organized groups targeting Hindu content.

Hoping to rally support, I posted about the issue in an internal Meta Employee Resource Group (ERG) for Hindus. What followed was a surprising and disheartening backlash. Many colleagues, most of them from Indian backgrounds, argued that the creator’s content was “pseudoscientific” and something their parents might share on WhatsApp, causing embarrassment. Instead of defending his work, they wanted it silenced, reflecting a deep discomfort with aspects of their own heritage. It was also a frightening insight into how the opinion of a small set of folks in the right place, could restrict information across the world.

This strengthened my resolve even further- I spent hours collecting data to demonstrate that the creator’s content didn’t violate Meta’s policies. I also noted examples of how other similar pages/creators (but with different ideological narratives) were allowed to thrive on the platform. I had to engage directly with product and policy teams, appealing to the leadership to consider the broader implications of such biases on the platform. It took me weeks of persistent effort, presenting data and appealing to product teams, to convince Meta’s leadership to manually review and reinstate the page. 

While the victory was meaningful, the struggle revealed a deeper problem: both the content moderation system and our own community were unprepared to stand up for Hindu narratives. And given my own capacity limitations, how many Hindu pages could I fight for? And why was Hindu information at the mercy of the efforts of a few people like me?

This incident left me with a sobering realization. If even content as benign as temple architecture could face such hurdles, the space for preserving and sharing the richness of Sanatana Dharma is far more fragile than we might think. It’s a call for all of us to do better, not just in creating our stories but in defending them.

Future Danger- AI Without Representation

With Meta’s move toward reducing manual content moderation and increasing reliance on crowdsourcing/user reporting and AI, the risks for Hindu content creators grow exponentially. The anti-Hindu bias of early AI platforms has already been a point of discussion.  Imagine, if manual intervention is tapered down as Meta has announced, the fate of many many Hindu creators and content would be left only to algorithms and ‘community’ moderation, and reverting these bans like I was able to (even with insider knowledge and persistence) will become next to impossible.  

Algorithms are not neutral. They are trained on existing data, which reflects the biases and narratives already dominating the digital space. If these narratives are overwhelmingly negative or distorted, the AI will learn and amplify them.

For example, consider the international media coverage of the Kumbh Mela. Instead of celebrating the spiritual and cultural grandeur of the world’s largest peaceful gathering, it is often reduced to images of ash-smeared sadhus with sensationalist captions. These are the perspectives that feed AI systems, shaping how future generations perceive Hinduism and Kumbh Mela –a “spectacle” not a sacred pilgrimage.

As Hindus, we cannot afford to remain complacent. AI is reshaping the world, and its impact will go far beyond content moderation. It will influence education, policymaking, and even collective memory. If we don’t act now, we risk seeing a version of Hinduism that is unrecognizable to us—a version that our children may grow up rejecting, not out of understanding but because of the biases they encounter online.

The Need for Organization and Action

One of the greatest challenges for Hindus has been our lack of organization and institutional representation, making it easy for anti-Hindu narratives to dominate unchecked.

The wars of the future will not be fought on traditional battlefields but in the realms of ideas, narratives, and digital spaces. In this new age, the pen—or rather the keyboard—is mightier than the sword. We must rise to the challenge by:

1. Creating High-Quality Content: We need more creators, writers, and educators producing content that celebrates the richness of Sanatana Dharma. From the sciences encoded in temple architecture to the philosophy of the Upanishads, there is so much to share with the world. Our ancestors’ art, architecture, science and writing have stood the test of time -that should be our model.
2. Building Support Systems: Platforms that provide legal, technical, and financial support to Hindu content creators can help counteract the challenges posed by biased algorithms or mass reporting. We need to build ecosystems that support, defend, cite and amplify authentic Dharmic work.
3. Grassroots Advocacy: Hindus must engage with tech companies to ensure fair representation in AI training data and content moderation policies. This includes forming think tanks and advocacy groups to make our voices heard, but can also be as simple as speaking up within the many many tech workplaces we work in. A lie left unchallenged-be it at school, workplace or social event is  on it way to becoming the truth. So at the very least, we can actively educate ourselves and defend Dharmic conversations on social media and in our own friend circles. 

Uttishthata Jagrata!

The digital age has given us tools of unprecedented power, but with that power comes responsibility. As Hindus, we are the inheritors of one of the world’s oldest and most profound traditions. It is our duty to ensure that this legacy is preserved and presented with the dignity it deserves.If we do not actively take charge of the narrative about our faith and heritage, we risk leaving it in the hands of algorithms trained on biased content and narratives perpetuated by institutions like the BBC or ‘scholars’ like Audrey Truschke. The outcome? A version of Hinduism stripped of its depth, beauty, and diversity—reduced to a caricature of casteism or a “cult of ash-smeared men”, unworthy of pride or preservation.

The future of Sanatana Dharma depends on what we do today. Will we rise to the occasion? Or will we let others define us for generations to come?

The choice is ours.