News · BFSI
India's cybersecurity agency has issued an urgent advisory identifying an AI-powered threat actor targeting NBFCs, digital lenders, payment aggregators and cryptocurrency exchanges in an ongoing campaign.
✦ AI SUMMARY 30-second read
- CERT-In designated the threat actor 'Crafty Falcon' and confirmed the campaign is active and ongoing, with affected entities re-targeted after incomplete remediation.
- Attackers used AI-driven tools — ARTEX, PentAGI, Cairn and CyberStrikeAI — to autonomously execute full attack chains, abuse payment APIs and move funds to mule accounts.
- CERT-In has recommended 24×7 monitoring with increased staffing, full forensic eradication of threats, cloud access restrictions and hardware-based storage of cryptocurrency wallet keys.
Why it matters: An autonomous AI threat actor exploiting payment and disbursement APIs at machine speed poses a systemic risk to India's digital financial infrastructure.
This summary was generated by AI. Editors may review and update it.
The Indian Computer Emergency Response Team (CERT-In) on Wednesday, November 4, issued an advisory warning that an AI-driven hacking campaign against entities in India's BFSI sector is active and escalating. The agency's Computer Security Incident Response Team for Finance (CSIRT-Fin) has designated the threat actor behind the campaign as 'Crafty Falcon'.
Non-banking financial companies, digital lending platforms, payment aggregators and cryptocurrency exchanges are among those affected. CERT-In said the threat actor used AI tools to survey targets and attack them by invoking the victims' own payment and disbursement APIs, all at machine speed.
The advisory builds on an earlier warning issued on October 9, when CERT-In disclosed observing multiple cyber attack campaigns targeting fintech firms that exploited security flaws in internet-facing applications, exposed APIs, privileged endpoints and stolen credentials to carry out fraudulent transactions to attacker-controlled mule accounts across multiple banks.
What was announced
CERT-In confirmed that affected fintech entities which patched vulnerabilities following the October 9 advisory have since been deliberately re-targeted due to incomplete remediation. The agency noted that the attackers timed the campaign to coincide with the festive period to exploit low-monitoring windows and delayed detection. 'All BFSI entities are strongly advised to treat this campaign as an immediate and continuing threat,' the advisory stated.
CSIRT-Fin's analysis identified four publicly available, AI-driven offensive security tools in active use: ARTEX, described as an open-source multi-agent AI penetration testing and command-and-control platform; PentAGI, an autonomous multi-AI agent penetration testing system; Cairn, an attack search engine; and the CyberStrikeAI framework. CERT-In said these tools execute complete attack chains autonomously at machine speed, generate large numbers of tailored attack scripts in rapid succession and adapt their payloads based on target responses. Attack infrastructure rotates IP addresses within hours to days.
Attack pattern
Crafty Falcon targeted AWS and Lambda cloud infrastructure used by affected organisations to execute commands and deploy malicious files. Non-production environments such as testing sites served as entry points, with attackers moving laterally to live production environments through shared credentials, tokens or network paths.
The actors reverse-engineered application codebases to develop tailored exploits and initiate unauthorised API calls using modified application packages, bypassing controls that assumed only legitimate client applications would call the APIs. Requests were routed through trusted third-party services such as content delivery networks and app-store review tools to mask malicious IP addresses.
Before making large-value transactions, the hackers conducted test transactions of Rs 1. Merchant settlement APIs of payment aggregators were abused to move funds, while cryptocurrency exchanges were targeted using stolen wallet keys. The attackers also quickly exploited newly disclosed flaws in the Node.js application and PHP.
Recommended safeguards
CERT-In has called for heightened 24×7 security monitoring with increased staffing and lowered alert thresholds during holidays and weekends. Entities are advised to carry out full forensic eradication and periodic threat hunting, given that incomplete remediation has led to re-attacks.
Cloud management and serverless execution should be restricted to approved principals, with centralised real-time monitoring of cloud audit logs and alerting on high-risk events such as remote command execution and creation of roles or access keys. UAT, staging and testing environments must be fully isolated from production, with no shared credentials, tokens, API keys, databases or direct network paths.
For cryptocurrency operations, CERT-In recommended storing wallet keys in hardware security modules or cold custody, requiring multi-signature approvals for withdrawals and monitoring withdrawal patterns for anomalies. The agency also advised organisations to monitor LLM API usage, noting that such traffic originating from servers or non-development hosts may signal an attack rig or planted operator console.
Separately, CERT-In this week issued new cybersecurity auditing guidelines for AI systems deployed by government agencies and private companies, requiring disclosure of reportable AI-related security incidents within six hours, alongside stricter requirements for risk assessment, monitoring and independent verification.
What happens next
CERT-In stated that detailed infrastructure fingerprints and detection rules have been shared with entities through restricted advisories. Given that the campaign is described as active and continuing, BFSI entities are expected to treat remediation as an ongoing process rather than a one-time exercise.
Elets will update this story as further developments emerge.
