Home / Resources / BFSI AI Visibility Study

We asked ChatGPT about 25 Indian BFSI brands. 17 were invisible.

When a first-time borrower asks AI which NBFC offers the best personal loan, or when a young professional asks which insurer has the best term plan, they expect an intelligent answer. Our study found that 17 of India's 25 leading BFSI brands were completely absent from those conversations — and 5 of the 8 that did appear were described incorrectly.

How we tested

We selected 25 brands spanning NBFCs, life and health insurers, and retail brokerage platforms — including Bajaj Finance, Muthoot Finance, IIFL Finance, Shriram Finance, Aditya Birla Capital, Tata Capital, Cholamandalam Finance, L&T Finance, Mahindra Finance, Sundaram Finance, InCred, Lendingkart, Max Life, HDFC Life, SBI Life, Star Health, Niva Bupa, Zerodha, Upstox, Groww, Paytm Money, 5paisa, Capital Float, Ugro Capital, and Incred Money.

For each brand we ran three prompt types across ChatGPT (GPT-4o), Gemini Advanced, and Perplexity:

• Category prompts — "best personal loan app in India for salaried professional", "which term insurance plan is recommended for a 32-year-old in India", "top discount brokers in India for long-term equity investing"
• Direct prompts — "what does [brand] do?", "is [brand] a good lender / insurer / broker?"
• Comparison prompts — "[brand] vs [competitor] — which is better for home loans?"

A brand scored visible if it appeared unprompted in at least one category answer. Accurate required no material error on product range, geography, or regulatory status.

What we found

• 17 of 25 brands never appeared in any category prompt across any engine. Their target customers, asking AI for exactly the service they provide, were consistently directed elsewhere.
• Of the 8 that did appear, 5 carried material errors — wrong product verticals, outdated AUM figures, regulatory body confused (IRDAI/SEBI/RBI mixed up), or descriptions conflated with similarly-named global entities.
• Only 3 of 25 were both visible and accurately described — a 12% pass rate in one of India's most competitive consumer categories.
• When AI engines did recommend BFSI products, they disproportionately cited large private banks (HDFC Bank, ICICI Bank, Kotak) — even when the query specifically asked about NBFCs or insurers. The category leaders offline are not the answers AI gives.

What the invisible 17 had in common

1. Compliance anxiety became content paralysis. SEBI, IRDAI, and RBI regulations are real, but most BFSI brands interpreted "regulated content" as "no content." Their digital presence consists almost entirely of disclaimer-heavy product pages and statutory filings — nothing an AI engine can meaningfully cite in a consumer-intent answer.
2. XBRL filings are not AI training material. A company's detailed annual report submitted to BSE/NSE doesn't feed AI answers the way a well-structured web page does. Regulatory disclosure ≠ AI-legible authority.
3. No owned original research. The brands with AI visibility published regular reports — interest rate analyses, insurance penetration data, wealth management outlooks. The invisible brands produced nothing an external journalist, researcher, or AI would cite.
4. Investor-relations content and customer-intent content are different things. Detailed quarterly presentations, MD&A writeups, and DRHP documents are designed for analysts, not for someone asking "which NBFC should I borrow from?" AI pulls from the latter type of content, which most BFSI brands haven't produced.

What to do about it

The fix is more tractable than most compliance teams assume. AI visibility in BFSI doesn't require relaxing regulatory standards — it requires publishing content that is already permitted but simply hasn't been written: comparison guides, original market data, decision frameworks, and expert commentary on category questions. The brands that do this consistently earn citation by default, because there's almost nothing else for AI to pull from. We've mapped the full method in our Complete GEO Guide.