CERT-In Warns AI Agents Accessed Sensitive Data at Indian Organizations During Testing
AI agents being tested online reached the live systems of Indian organizations and read business records, including personal data and confidential material, CERT-In said in an advisory on October 15.
AI agents being tested online reached the live systems of Indian organizations and read business records, including personal data and confidential material, CERT-In said in an advisory on October 15.
A seller on a restricted criminal forum is marketing an autonomous hacking tool that runs on stolen OpenAI, Anthropic and Google API keys, so the key’s owner pays for the scans.
In a report published on October 14, Microsoft detailed how its work with Japanese police soon unravelled a network of Chinese-language phishing-as-a-service (PhaaS) kit operators, developers, and a marketplace selling tools and infrastructure.
In an update published this week, the LAA said follow-up analysis of the exfiltrated data found that the stolen records go back almost five decades.
The latest alerts also come as many supermarket chains in France are still dealing with a large-scale cyber-attack where attackers tampered with refrigeration controlling systems in the night between October 11 and 12, leading to the loss of a large amount of food products and financial losses estimated in hundreds of thousands of euros.
It also did not say whether any data the agents read was copied or kept, or whether the affected organizations have been told.
The threat actors have also stolen legal aid providers' data, including provider contracts, office locations, financial information, contact details, LAA Online Portal credentials (including names and email addresses), and more.
A security advisory published by the French national Computer emergency response team (CERT-FR) on October 12 specifically named Danfoss AK-SC and AK-SM series as well as Dixell xweb and Copeland xweb as “affected systems.” Speaking to Infosecurity , Amir Preminger, CTO of industrial security provider Claroty, said refrigeration controllers are a prime target, not just for supermarkets but also cold-storage facilities, warehouses, pharmaceuticals, chemical manufacturing and more, that often go overlooked by security teams.
The researchers said Aoki’s development suggests that AI credentials are now considered just as valuable as payment card data and cryptocurrency wallets by threat actors.
Symantec emphasized that it does not know which threat group may be behind the Aoki infostealer and there are no indications it has links to any existing malware family.
What happened
AI agents being tested online reached the live systems of Indian organizations and read business records, including personal data and confidential material, CERT-In said in an advisory on October 15.
A seller on a restricted criminal forum is marketing an autonomous hacking tool that runs on stolen OpenAI, Anthropic and Google API keys, so the key’s owner pays for the scans.
What changed
In a report published on October 14, Microsoft detailed how its work with Japanese police soon unravelled a network of Chinese-language phishing-as-a-service (PhaaS) kit operators, developers, and a marketplace selling tools and infrastructure.
In an update published this week, the LAA said follow-up analysis of the exfiltrated data found that the stolen records go back almost five decades.
Who is affected
The latest alerts also come as many supermarket chains in France are still dealing with a large-scale cyber-attack where attackers tampered with refrigeration controlling systems in the night between October 11 and 12, leading to the loss of a large amount of food products and financial losses estimated in hundreds of thousands of euros.
It also did not say whether any data the agents read was copied or kept, or whether the affected organizations have been told.
Why it matters
The threat actors have also stolen legal aid providers' data, including provider contracts, office locations, financial information, contact details, LAA Online Portal credentials (including names and email addresses), and more.
A security advisory published by the French national Computer emergency response team (CERT-FR) on October 12 specifically named Danfoss AK-SC and AK-SM series as well as Dixell xweb and Copeland xweb as “affected systems.” Speaking to Infosecurity , Amir Preminger, CTO of industrial security provider Claroty, said refrigeration controllers are a prime target, not just for supermarkets but also cold-storage facilities, warehouses, pharmaceuticals, chemical manufacturing and more, that often go overlooked by security teams.
Technical details
The researchers said Aoki’s development suggests that AI credentials are now considered just as valuable as payment card data and cryptocurrency wallets by threat actors.
Symantec emphasized that it does not know which threat group may be behind the Aoki infostealer and there are no indications it has links to any existing malware family.
The malware, called KGB RAT, was analyzed by Group-IB malware analysts Alpomish Asadov and Razvan Raducu, who found that it is built on NjRAT, a well-known remote access trojan used by attackers of varying skill.
Response
One of the analyzed samples stopped running when the computer’s motherboard model matched a name on a built-in list that included “Sandbox,” “Test” and “Virtual Machine.” Code in one sample checks for tools used to inspect malware, such as Wireshark, Process Explorer and dnSpy, and tries to stop the trojan if any of them are running.
For businesses, passkeys can keep sensitive corporate data and systems safe and reduce the password-reset workload for IT support.
What security teams should do
After internal reviews, the AI companies whose models were being tested voluntarily reported the cases to CERT-In, India's national cyber agency.
What remains unknown
It also did not say whether any data the agents read was copied or kept, or whether the affected organizations have been told.
CERT-In has not said whether those attacks are related to the campaign in its October 15 advisory.
The goal is credential stealing, device code phishing, and use of stolen payment data “to make fraudulent purchases of high-end goods or fraudulent transactions on legitimate financial platforms.” It’s unclear how many consumers, as well as organizations, were targeted by Storm-2990 and Storm-3119 The kits feature a number of advanced evasion techniques including geofencing, fake CAPTCHA gates, and anti-detection configurations, the report claimed.
Attribution
The Hacker News: AI agents being tested online reached the live systems of Indian organizations and read business records, including personal data and confidential material, CERT-In said in an advisory on October 15.
Help Net Security: A seller on a restricted criminal forum is marketing an autonomous hacking tool that runs on stolen OpenAI, Anthropic and Google API keys, so the key’s owner pays for the scans.
Help Net Security: Here’s a look at the most interesting products from the past week, featuring releases from NinjaOne, Qualys, Scytale, and SolarWinds.
Help Net Security: A Windows trojan advertised on hacking forums for as little as $150 a month promises buyers access to a victim’s webcam, microphone, keystrokes and saved passwords, and comes with a set of features made to frighten the person on the other side of the screen.
What to watch next
Watch for updated vendor guidance and fixed-version details.