A security flaw in OpenAI’s ChatGPT application programming interface could be used to initiate a distributed denial-of-service attack on websites, according to a researcher. The discovery was made by Benjamin Flesch, a security researcher in Germany, who detailed the vulnerability and how it could be exploited on GitHub.
According to Flesch, the flaw lies in the API’s handling of HTTP POST requests to the /backend-api/attributions endpoint. The endpoint allows a list of hyperlinks to be provided through the “urls” parameter. The problem arises from an absence of limits on the number of hyperlinks that can be included in a single request, so attackers can easily flood requests with urls via the API.
Read more…
Source: Silicone Angle News
Related:
- Ransomware gangs are using virtual machines to disguise their attacks
June 24, 2021
Cyber criminals are increasingly using virtual machines to compromise networks with ransomware. By using virtual machines as part of the process, ransomware attackers are able to conduct their activity with additional subtlety, because running the payload within a virtual environment reduces the chances of the activity being discovered – until it’s too late and the ransomware ...
- Malicious spam campaigns delivering banking Trojans
June 24, 2021
In mid-March 2021, we observed two new spam campaigns. The messages in both cases were written in English and contained ZIP attachments or links to ZIP files. Further research revealed that both campaigns ultimately aimed to distribute banking Trojans. The payload in most cases was IcedID (Trojan-Banker.Win32.IcedID), but we have also seen a few QBot ...
- Critical Palo Alto Cyber-Defense Bug Allows Remote ‘War Room’ Access
June 23, 2021
A critical security bug in Palo Alto Networks’ Cortex XSOAR could allow remote attackers to run commands and automations in the Cortex XSOAR War Room and to take other actions on the platform, without having to log in. Found internally by Palo Alto, the bug (CVE-2021-3044) is an improper-authorization vulnerability that “enables a remote unauthenticated attacker ...
- REvil Ransomware Code Ripped Off by Rivals
June 23, 2021
They say imitation is the sincerest form of flattery: The LV ransomware, a strain that cropped up just this spring, turns out to be based on what is most likely pirated REvil ransomware code, according to researchers. A malware analysis of LV from Secureworks Counter Threat Unit (CTU) found that its operators (which it calls Gold ...
- How to confuse antimalware neural networks. Adversarial attacks and protection
June 23, 2021
Nowadays, cybersecurity companies implement a variety of methods to discover new, previously unknown malware files. Machine learning (ML) is a powerful and widely used approach for this task. At Kaspersky we have a number of complex ML models based on different file features, including models for static and dynamic detection, for processing sandbox logs and ...
- NukeSped Copies Fileless Code From Bundlore, Leaves It Unused
June 22, 2021
While investigating samples of NukeSped, a remote access trojan (RAT), Trend Micro came across several Bundlore adware samples using the same fileless routine that was spotted in NukeSped. The backdoor has been attributed to the cybercriminal group Lazarus, which has been active since at least 2014. There are multiple variants of NukeSped, which is designed ...