The Reserve Bank of India (RBI) has consolidated its cyber security rules for non-bank payment system operators into a single master direction, issued on November 5. These are companies that run payment systems without being banks. They include the National Payments Corporation of India (NPCI), which operates UPI, as well as card networks, payment aggregators and prepaid wallet issuers.
A master direction is RBI’s consolidated rulebook on a subject. This direction combines the rules RBI issued in 2024 with earlier rules on ATM switch providers and annual system audits. RBI said the aim is to “provide a single reference point for the entities”.
For users, the direction lists the fraud safeguards payment apps must offer. These include a 12-hour block on online payments after a user changes their registered mobile number or email address, a block on payments while remote access apps are running, and an option to report fraudulent transactions.
What were the 2024 rules?
In July 2024, RBI issued master directions on cyber resilience and digital payment security for non-bank payment companies. The directions made company boards responsible for cyber risk, required companies to report cyber incidents to RBI, and introduced fraud safeguards for payment apps. Large companies had to comply by April 2025, while medium companies had until April 2026.
What’s new?
- ATM switch providers: The direction incorporates RBI’s 2019 security controls for third-party ATM switch providers. An ATM switch routes a card transaction from an ATM to the customer’s bank for approval. The rules cover switch providers that serve white label ATM operators, which are non-bank companies that run ATMs.
- System audits: It incorporates RBI’s 2020 rules on what an annual system audit, an independent review of a company’s IT systems, must cover.
- Repealed circulars: RBI has repealed six circulars on system audits and ATM switch controls issued between 2009 and 2020.
- Timelines: The direction applies immediately to large and medium payment companies. Small payment companies have until April 1, 2028, to comply.
What safeguards do users get?
- 12-hour cooling period: When a user changes the mobile number or email address linked to a payment instrument, “there shall be a cooling period of minimum 12 hours before allowing any payment transaction through online modes / channels.”
- Remote access block: Payment apps must detect remote access apps, “to the extent possible”, and block access to the payment app “while the remote access is live”. Remote access apps let another person view and control a phone.
- Device binding: Payment apps must be linked to the user’s phone and SIM card so that they work only on that device. If the app remains unused beyond a specified period, the device binding must be done again.
- Fraud reporting option: Apps and websites must let users “identify / mark a fraudulent transaction for seamless and immediate notification to the issuer of payment instrument.”
- Clearer alerts: Alerts for online payments must name the merchant and the amount. They must not show the payment gateway or aggregator in place of the merchant. Fund transfer alerts must show the amount debited and the beneficiary’s name, as recorded by the beneficiary’s bank. When an alert contains a one-time password (OTP), the OTP must appear “at the end of the notification message”. The message must also refer to the specific transaction.
- Masked details: Alerts must mask account and card numbers, “to the extent possible”.
- Login controls: Apps must log users out after a period of inactivity. Where applicable, they must block access after a specified number of failed login attempts and notify users of each failed attempt “immediately”.
- Multi-factor authentication: Every electronic debit, including cash withdrawals, must pass more than one check, such as a PIN and an OTP, “except where explicitly permitted / relaxed.”
- 24×7 fraud desk: Companies must operate a manned facility “on 24x7x365 basis” to resolve fraud complaints and respond to law enforcement agencies.
- Wallet safeguards: Prepaid wallet issuers must set a cooling period before users can transfer or withdraw money newly loaded onto a wallet. RBI encourages them to send OTPs and alerts to users “in a language of their choice, including vernacular languages.”
Who does it apply to?
RBI groups payment companies into three categories:
- Large: NPCI, NPCI Bharat BillPay Limited, card networks, payment aggregators, large prepaid wallet issuers, non-bank ATM networks and white label ATM operators. Payment aggregators let online merchants accept payments. This category also includes:
- The Clearing Corporation of India Limited (CCIL), which settles trades in government securities and foreign exchange.
- Bharat Bill Payment operating units, which handle bill payments through the Bharat Bill Payment System.
- Trade Receivables Discounting System (TReDS) platforms, which help small businesses receive early payments against invoices.
- Medium: Inbound cross-border money transfer operators and medium prepaid wallet issuers.
- Small: Small prepaid wallet issuers and instant money transfer operators.
Payment companies must also ensure that unregulated partners in their payment chain, such as payment gateways, vendors and ATM switch providers, follow the rules, “subject to mutual agreement”. Each company must have a board-approved policy for this.
What must payment companies do?
- Board oversight: The board is responsible for cyber risk. It can delegate primary oversight to a board committee led by a member with experience in information or cyber security. The committee must meet at least once a quarter.
- Incident reporting: Companies must report cyber attacks, critical system outages, internal fraud and settlement delays to RBI “within 6 hours of detection”. They must also report cyber security incidents to the Indian Computer Emergency Response Team (CERT-In), the government agency responsible for handling cyber security incidents.
- Fraud monitoring: Companies must operate a “real-time / near real-time fraud monitoring solution” and a security operations centre to monitor their networks and systems for attacks.
- Risk-based alerts: Companies must help their member banks and partners set up alerts for warning signs. These include unusual locations or IP addresses and behavioural biometrics, such as changes in how a user types or swipes. They also include payments to UPI IDs, wallets or mobile numbers already linked to fraud.
- Data localisation: Companies must follow rules on where payment data is stored and processed and ensure that their vendors do the same. Their cloud policies must also cover data localisation.
- Disaster recovery: Companies must maintain a backup site that can process payments if the main data centre fails. The backup site must be in a different seismic zone, and companies must conduct recovery drills at least once every six months.
Companies must also appoint a senior security head, such as a Chief Information Security Officer (CISO), and use anti-phishing services to take down fake websites and apps. They must retain records of system activity for at least five years and secure access to the source code of critical software purchased from vendors.
What else does the direction require?
- Annual audits: Payment companies must have their IT systems independently audited every year. The audit must cover 20 areas, ranging from data security to mobile apps. Companies following the April-March financial year must submit the report to RBI by June 1 and fix any problems within three months.
- ATM switch providers: White label ATM operators must ensure that their ATM switch providers follow RBI’s security rules through contractual agreements. The providers must block unapproved software, cut off remote access to the network hosting the ATM switch and establish a security operations centre. They must also maintain protection against attacks that crash systems by flooding them with traffic.
What else is RBI doing on cyber security?
- Advisory to banks: RBI issued a confidential advisory dated November 3, asking banks to tighten security on the APIs that NBFCs and fintechs use to move money through banks. APIs are interfaces through which apps exchange data. The advisory follows a rise in AI-assisted cyber attacks on the financial sector. On November 4, CERT-In identified one threat actor, “CRAFTY FALCON”, which it said used Rs 1 transfers to test fund flows before making larger transactions.
- Audits and controls: Under the advisory, banks must get NBFCs using their APIs audited by CERT-In-empanelled auditors. Existing NBFC clients have until February 3, 2027, to comply. Banks must also immediately set per-transaction and daily limits. By December 3, 2026, they must enforce beneficiary whitelisting, which allows payouts only to pre-registered accounts, or maker-checker approval, which requires a second person or channel to confirm a debit.
- Kill switch for customers: RBI is examining a “kill switch” that would let customers instantly block all debit transactions from their accounts, according to its 2025-26 annual report, Free Press Journal reported. The proposal has not been finalised.
- Kill switch for AI systems: RBI’s draft AI guidelines would require banks to be able to override, suspend or switch off any AI system they use.
Also Read: