Fight Against Fake News: Meet the WhatsApp Hero
Ever feel like your family chat in India is a nonstop newsfeed of wild rumors? From a “water crisis in South Africa” to a Bollywood star’s supposed death, the gossip can blow up faster than a Bollywood blockbuster. That’s exactly how Tarunima Prabhakar, co‑founder of the India‑based tech firm Tattle, realized she needed a solution.
What did she build?
- A web‑based tool that checks messages against verified data from fact‑checking sites and reputable news outlets.
- Machine‑learning magic that automatically flags or confirms information, saving users the time and hassle of manual research.
- Free access for students, researchers, journalists, and academics—because we all need to stay sane in the digital age.
Why WhatsApp? Why India?
WhatsApp isn’t just a ‘messenger.’ With over 2 billion users worldwide and a staggering half‑billion in India alone, it holds power that rivals Facebook and Twitter. Yet while those giants are under fire for misinformation, WhatsApp remains largely ignored. The platform rolled out story‑forward controls in 2018 after tragic riots sparked by fake rumors. Still, those measures don’t match the nuance needed for India’s diverse languages and local contexts.
Key Challenges & Real‑World Impact
- Language barometers – Current tools aren’t built to understand Indian tongues.
- Local grit – Social media companies often ignore the cultural, historical, economic, and political nuances of their users.
- Polarization & violence – Misinfo can fuel real‑world tension, as experts warn.
According to Pierre François Docquir, head of media freedom at Article 19, “These platforms don’t listen to local communities. The lack of moderators who understand local dialects and events is a recipe for disaster.”
What Has Tattle Done?
- Uses AI to crowdsource fact‑checking, ensuring the tool stays sharp.
- Partners with civil‑society groups to train users and build trust.
- Creates a scalable framework that can be adapted for other Asian languages and cultures.
So, next time you see a shocking claim pop up in your WhatsApp group, remember Tarunima’s tool is just a click away—ready to separate the truth from the nonsense with a dash of tech and a pinch of local insight.
Local initiatives vital
Online Hate Speech: The Silent Emergency in Asian Tech
Everyone knows that the internet can be a wild west, but when it turns into a breeding ground for hate, things get seriously messy. Across Asia, tech giants quietly shrug at the rampant toxicity and are barely scraping the surface when it comes to moderating content in local tongues.
What Really Happened in Myanmar
In 2017, the Rohingya community went through a brutal crackdown, and Facebook was in the cross‑fire. UN rights investigators traced a chain of hateful posts that escalated the violence. Facebook, putting on a calm front, talked about fighting misinformation and “investing in Burmese‑speaking talent.” But the damage was already halfway done.
Indonesia’s “Epic Saga” of Hate
- Hate isn’t just limited to the obvious—religious minorities, racial groups, and LGBTQ+ folks all become easy targets.
- Bots and paid trolls flood the network, feeding disinformation that splits communities further apart.
- Article 19’s June report named this as “significant hate speech” spamming our feeds.
- “Social media companies must team up with local players,” wrote researcher Sherly Haristya.
Enter Mafindo: The Avengers of Fact‑Checking
Meet Mafindo – short for Masyarakat Anti Fitnah Indonesia. With a backbone of Google’s support, this non‑profit is training everyone from school kids to stay‑at‑home moms to sniff out misinformation.
What they do:
- Hands‑on sessions on reverse image searches, video metadata, and geolocation techniques.
- A dedicated fact‑checking squad bolstered by citizen volunteers.
- So far, they’ve debunked 8,550 hoaxes – that’s like opening a book and finding every single word wrong!
Kalimasada: The WhatsApp Whisperer
Before the 2019 elections, Mafindo launched Kalimasada, a chatbot that answers in Bahasa. It’s accessible via WhatsApp and has already captured around 37,000 curious users. Think of it as the “Siri” for fighting fake news.
Why the Elderly Need a Hand
Last but not least, the senior citizens are especially vulnerable. Santi Indra Astuti, Mafindo’s president, knows the drill:
“The older generation often falls prey to fake news because they’re not tech‑savvy and can’t navigate the internet easily.”
Her team teaches:
- How to use social media safely.
- Protecting personal data.
- Questioning trending topics – from COVID‑19 vaccine hype to 2020 election rumors.
Bottom line? If we want a healthier online space, we need tech to step up along with grassroots initiatives. The digital world is fast and fierce—so pretty sure we shouldn’t let it turn into a toxic playground.
Abuse detection challenges
Asia Tightens Social‑Media Rules, but Local Hate Speech Still Finds a Way to Sneak In
Governments across Asia have just rolled out a new set of rules that put the brakes on the hottest social‑media platforms. Think “no‑flooding” of certain message types, “quick‑delete” mandates for content that the regulators deem objectionable, and a whole lot of legal paperwork for platforms that are already overworked. But if you look closely, the most nasty troll‑language—that’s written in the local dialects—keeps slipping past the filters.
Meet Prabhakar, the Back‑Ground “Chisel” Ninja
While the laws go top‑down, tech‑savvy folks are fighting the battle from the ground up. One of those players is Prabhakar of Tattle, who’s built a tool called Uli (Tamil for “chisel”). This isn’t your ordinary screenshot‑capturer; it actually “chisels” away the ugly bits from user feeds. Uli works in English, Tamil and Hindi—so if you’re sending a meme in an Indian dialect, the tool will politely hide it.
How Uli Gets Smarter
- Crowdsourced Lexicon – Tattle’s team has collected a growing list of offensive words from the internet. The “real” chatroom or social‑media space becomes a cleaner place because all posts containing those flagged words get blurred.
- Machine‑Learning & Pattern Recognition – Every user can add their own slurs to the list, and the tool learns to spot new slurs by comparing patterns instead of having everyone manually scroll through posts.
- Bottom‑Up Moderation – Unlike the usual “global…platform‑wide” approach, Uli acts directly on a user’s feed. Think of it as a personal “digital bouncer” who screens strangers before they even cross the door.
- Future Vision – Uli’s roadmap includes detecting non‑textual abuse such as memes, images and even videos that are basically “animated war cries” in disguise.
Singapore’s Own Tale: Empathly, the Spell‑Check of Hate
Enter Empathly – a start‑up born from two university students. Like that “spell check” you’re used to on your phone, Empathly will flag the offensive words before they become offense.
Why Empathly Gone “Local” Matters
- Multilingual Mastery – English, Hokkien, Cantonese, Malay and Singaporean English (a.k.a. Singlish) are all covered. The nerds know why “shiok” might be a curse word for some
- Culture‑Smart Flagging – In a hyper‑connected region, local understandings of offensive content matter. Empathly is built inside Singapore’s own linguistic ecosystem.
- Business‑Friendly – It’s aimed at corporate giants who want to stay on the right side of the law with zero hassle. Think “no‑poop” compliance, but for words.
— Timothy Liau, founder and CEO of Empathly
“We’ve seen the harm that hate speech can cause. Big Tech bills its users in English and focuses on English speaking markets. So there’s a niche for local intervention.”
Why It Still Matters
Policy is tightening, but the fight against local hate speech isn’t a zero‑sum game. “Abuse detection is very challenging,” Prabhakar acknowledges—especially when the slang changes faster than a rabbit in a hoover. The deeper the local nuances, the tougher the battle—and the sweeter the victory when you get it right.
In conclusion: Below the global regulatory radar, a world of dedicated local tech gadgets is emerging. Whether it’s a chisel that chisels a bad post out of a feed, or a spell‑check that stops offended comments before they hit the “send” button, the fight for a kinder online space continues—one correctly flagged emoji at a time.
