CERT-In Advisory Flags Payment Fraud and Agentic AI Attacks By Elets News Network - 21 November 2026

Close-up of JavaScript code in a text editor on a laptop screen, illustrating CERT-In's payment workflow advisory

The latest CERT-In advisory, CIAD-2026-0062, issued on 20 November 2026 by the Indian Computer Emergency Response Team, warns that unauthorised payments made by tampering with business records and approval workflows have spread beyond banks and fintechs. Rated critical, it says technology service providers supporting government-linked payments and vendor payout platforms are now hit, alongside AI-driven agentic attacks.

Titled “Abuse of Business Payment Workflows and Use of AI-Driven Agentic Offensive Frameworks across Sectors”, the advisory follows recent CERT-In alerts, including one earlier this month on the Crafty Falcon AI hacking campaign against fintechs and NBFCs.


How the CERT-In payment workflow attacks work

CERT-In said attackers are not just breaking into payment APIs. They change the records that decide whether a payment should go through, then let legitimate systems do the rest. Methods seen in reported incidents include:

  • Misuse of business applications and workflows operated by service providers.
  • Alteration of business records or approval relationships that determine payments.
  • Use of trusted systems and approval procedures, so transactions do not raise immediate suspicion.
  • Exploitation of gaps in independent verification, reconciliation and review of record changes.

“A transaction may pass the usual application and approval checks even when the records used to initiate it have been changed without authorisation,” the advisory said. CERT-In did not disclose how much money was lost or which organisations were affected.


What are AI-driven agentic offensive frameworks?

In separate threat hunting, CERT-In found AI-driven agentic offensive frameworks (AOFs) used against enterprises, service providers, government bodies and research institutions. These tools automate reconnaissance, rapidly prepare exploits for known vulnerabilities, run multiple agents in parallel and adapt when blocked, using valid credentials and trust relationships to go deeper.

CERT-In said AOFs cut manual effort and shorten the gap between reconnaissance and attack. It follows an earlier CERT-In alert on AI agents reading Indian firms’ data undetected.

What the CERT-In advisory asks organisations to do

Beyond measures in its earlier advisories and its May 2026 blueprint on AI-assisted exploitation, CERT-In set out targeted steps:

  • Government bodies and service owners: map provider accounts and integrations that can change payment records or approve transactions; reconcile payment instructions against independent records that the same provider environment cannot alter; and require independent verification for changes to beneficiaries and approvals.
  • Technology service providers, ITES firms and payout platforms: separate client environments, restrict administrative and control-plane privileges, and monitor record changes alongside privileged activity and transactions.
  • Organisations exposed to AOFs: review internet-facing applications, identities and credentials, and check logs for unusual reconnaissance and outbound data transfers.

Incidents must be reported to CERT-In within six hours under its April 2022 Directions, the same window set in its AI security audit guidelines. Unauthorised transactions should also be reported on the National Cyber Crime Reporting Portal or helpline 1930. The full text is on the CERT-In advisory page.

Why it matters for CIOs and CISOs

The advisory shifts attention from the payment gateway to the business process. Multiple approvals inside one provider-controlled system may not protect an organisation if that environment is compromised, CERT-In noted. For enterprises scaling agentic AI, it adds weight to questions of accountability when AI agents act inside business systems.

Key takeaways

  • CERT-In advisory CIAD-2026-0062, issued on 20 November 2026, is rated critical.
  • Payment fraud via altered records and approvals now affects government-linked payment providers and payout platforms, not just BFSI.
  • Agentic AI attack frameworks are automating reconnaissance and intrusion across sectors.

FAQ

What did CERT-In warn about on 20 November 2026?

CERT-In warned that attackers are altering business records and approval workflows to push unauthorised payments, and that AI-driven agentic offensive frameworks are being used in cyberattacks across sectors.

Who should act on the CERT-In advisory?

Government organisations that rely on technology service providers, the providers and ITES firms themselves, non-financial payout platforms, and any organisation exposed to AI-driven intrusions.

Featured image: Code in a text editor on a laptop screen. Photo: Marc Mueller via Unsplash / Wikimedia Commons (CC0).

Be a part of Elets Collaborative Initiatives. Join Us for Upcoming Events and explore business opportunities. Like us on Facebook , connect with us on LinkedIn and follow us on Twitter.

"Exciting news! Elets technomedia is now on WhatsApp Channels Subscribe today by clicking the link and stay updated with the latest insights!" Click here!

Related artificial intelligence