Source attribution and framing analysis on every story. Built for India's 900M readers.
Parliament passes amendment after extended debate; opposition stages walkout citing procedural concerns.
The bill passed 287 to 142 after four hours of debate. Three opposition parties left the chamber before the vote.
We surface the phrase so readers weigh it, not absorb it.
Scans for loaded language, missing attribution, and sensational framing. The full pipeline also runs source verification and native Indic analysis.
Government slams critics as opposition desperately scrambles to defend its embarrassing record on inflation.
Heavy framing. Multiple loaded signals.
Reliable content from authentic Indian and international sources. Every story delivered as a 60-word summary with full source verification.
Recommendation-first, not publisher-first. The feed learns what you read, skip, share, and return to — then recommends source-attributed content in your language.
Most misinformation starts as selective framing, not fake news. Our pipeline runs four checks between publish and feed.
Stories pulled from authentic Indian and international publishers. ~10,000 articles a day across Hindi, English, and Bangla.
Each story read in its original Indic language. Bias lives in idioms and honorifics — it's lost when translated to English first.
Every claim tied back to its source. Loaded language, one-sided framing, and missing attribution flagged before publishing.
Each story condensed to a 60-word summary with the source visible. Scan, click to verify, decide for yourself.
The problem isn't access. It's provenance, framing, and language. Here's how we compare with the platforms readers use today.
900M internet users. Supply of unverified, AI-generated, emotionally framed content has outscaled every safeguard.
Closed sharing strips source attribution. By the time it reaches the reader, the origin is gone.
Synthetic articles and fabricated quotes now flood Indic-language feeds. Detection lags creation.
Most fact-checking tools assume English. Bias in Hindi or Bangla idioms stays invisible.
It lives in six high-frequency categories most apps treat as sidebars. We treat them as the main wedge.
Festival timing and life-event guidance, with named source astrologers.
Film, music, OTT — with paid-placement detection on every review.
Live scores, IPL coverage, player news from verified beat reporters.
Temple events and sacred calendars, attributed to recognised bodies.
Markets, policy, personal finance with source pills on every claim.
National, state, and constituency coverage with framing flags.
“Bias in Hindi or Bangla cannot be detected by translating to English first. It lives in idioms, honorifics, and local framing.”
We don't claim to determine the truth. We show what was reported, who reported it, and how it was framed.
Hindi, Bangla, and English today; 19 more on the roadmap. Built on AI4Bharat embeddings with a proprietary Indic news SLM in development.
Not publisher-first. The feed learns what you read, skip, share, listen to, and return to — across every category.
AI-generated audio and AI anchor short-form video for verified stories, rolling out across Indian languages.
Regional journalists, domain experts, and institutions publishing through the same provenance layer — revenue-shared as trusted inventory scales.
Two ends of the same problem. Carnegie Mellon on the model side; Government of India on distribution. Eighteen months together, always in person.
Built the original AI pipeline. Owns model architecture, the neutralisation eval system, and the Indic SLM roadmap.
Owns strategy, growth, distribution, partnerships, editorial. Also founder of Green Governance India.
Owns the application stack, infra, ingestion, and ML serving. Works directly with Umesh on AI integration.
Owns content strategy, editorial standards, and the verified creator program.
Owns revenue, partnerships, GTM. Leads B2B verification pilots and institutional channels.
Over the next twelve months we extend verified content into audio and video, and open the platform to verified journalists, creators, and institutions.
Hindi, English, Bangla. Every story scannable with full source attribution.
Verified stories listenable in Indian languages. Powered by Sarvam APIs.
Same trusted summary in 30 seconds. Provenance preserved across formats.
Verified regional journalists publishing through our provenance layer.
Verified creators across ABCDEP earning through a revenue share model.
Government departments and civic bodies reaching citizens in their language.
Free. iOS, Android, Indus AppStore. 10,000+ stories a day with source attribution and framing analysis.
