Yes Bank API changes are delaying fintech payouts in India
Yes Bank's new defences against AI-driven attacks are slowing settlements for payment firms that rely on its banking APIs, according to industry sources.

Yes Bank is tightening its defences against AI-driven cyberattacks, and the fintech companies that plug into its systems are the ones feeling it first. Industry sources say the upgrades are disrupting workflows and delaying payouts at payment service providers.
One fintech executive said problems began after Yes Bank sent an email saying it was "implementing enhanced precautionary measures to safeguard" transactions. The company was later told the bank was changing its banking APIs, the connections that let businesses move money programmatically, and that this was delaying settlements. Some businesses relying on those APIs are now seeing slower payment processing and settlement times.
Another insider linked the trouble to directions the Reserve Bank of India (RBI) issued to banks and financial institutions to strengthen cybersecurity. According to that source, they followed the October data leak at the State Bank of India (SBI). The RBI is also said to have told lenders to make sure customer funds are not hit by the growing wave of AI-assisted attacks. Yes Bank has been asked for comment.
Why are Indian fintech withdrawals delayed?
Yes Bank is not the only place where money is moving slowly. Coinbase India paused all withdrawals last week before partly resuming them, citing an outage at its banking partner. CoinDCX said upgrades at "most" of its partner banks were slowing INR withdrawals, and Zebpay blamed a technical disruption at its banking partner. Razorpay told customers it would temporarily delay merchant settlements because of a technical issue at a partner bank, then restored the service hours later.
What did CERT-In warn about?
On the same day, the national cybersecurity agency CERT-In issued an advisory on unauthorised transactions at technology service providers handling government-linked payments and vendor payouts. Attackers altered business records or approval relationships, then used legitimate payment processes to approve fraudulent transactions. CERT-In said they misused business applications and workflows rather than only exploiting payment APIs, and did not say how much money was taken.
The agency listed AI-assisted techniques, including automated reconnaissance, rapid tweaking of exploits for known flaws, multiple automated agents working in parallel, and use of valid credentials and admin functions. It told government entities to restrict accounts that can approve payments, urged banks to reconcile provider data against authoritative records, and asked fintechs to watch for unusual activity and changes to payment records.
Apps worth a look
Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Talk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
Open EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
Pick who it is for, the occasion and their personality. The search engine reads product descriptions from local shops and ranks what fits. Free, no sign-up.
More from this day
See all →
Meta Muse Chrome extension: AI agent now lives in a sidebar

Medical microrobots steered by magnets could replace incisions

How to remove your personal data from Google with Results about you

Gmail Shielded Email gives you a throwaway address in one tap
More in: Business
See all →
Chinese 2-cent pajama sale draws 2 million orders, then vanishes

Take-Two lets a GTA trademark lapse in Russia after 18 months

SpaceX Project Saturn: Lenders Don't Want Nvidia Chips in Orbit
