AI debt collection platform exposure leaves millions of accounts and call logs open

Executive Briefing
“Autonomous debt collection platform CreditNirvana exposed 6.6 million accounts and hundreds...”
Key Takeaways
- 1.Independent security researcher Eaton Zveare discovered two vulnerabilities in the CreditNirvana platform owned by Indian firm Perfios, working alongside India's Computer Emergency Response Team.
- 2.The incident exposed 6.6 million accounts and approximately 315,000 call recordings, including credit scores, legal documents, emails, text messages, and case management records.
- 3.CERT-In categorized one vulnerability as high severity and the other as critical, potentially allowing attackers to hijack automated tasks or impersonate administrative accounts.
- 4.The systems were patched within a single day after being addressed in June before public disclosure followed in October.
Autonomous debt collection platform CreditNirvana exposed 6.6 million accounts and hundreds of thousands of call recordings due to two security flaws. The incident highlights mounting data risks as financial institutions integrate automated intelligence into sensitive collection workflows.
Incident Details & Technical Context
Independent security researcher Eaton Zveare discovered two vulnerabilities in the CreditNirvana platform owned by Indian firm Perfios, working alongside India's Computer Emergency Response Team.
The incident exposed 6.6 million accounts and approximately 315,000 call recordings, including credit scores, legal documents, emails, text messages, and case management records.
CERT-In categorized one vulnerability as high severity and the other as critical, potentially allowing attackers to hijack automated tasks or impersonate administrative accounts.
The systems were patched within a single day after being addressed in June before public disclosure followed in October.
Risk Assessment & Security Implications
Integrating autonomous models into debt workflows expands the attack surface, allowing a single software flaw to compromise massive troves of sensitive personal financial history.
This incident illustrates how automated tooling not only stores sensitive information but also directs operational campaigns, turning technical weaknesses into business process manipulation risks.
Mitigation & Factors to Watch
The exact duration that these vulnerabilities remained present within the infrastructure prior to discovery remains unquantified.
There is no definitive confirmation regarding whether malicious actors actively exploited these flaws before patches were applied.







