Jailbreaking Leaves a Trace: Understanding and Detecting Jailbreak Attacks from Internal Representations of Large Language Models

Jailbreaking large language models LLMs has emerged as a critical security challenge with the widespread deployment of conversational AI systems. Adversarial users exploit these models through carefully crafted prompts to elicit restricted or unsafe outputs, a phenomenon commonly referred to as...

CVE-2025-62615

AutoGPT's ReadRSSFeedBlock contained an SSRF flaw due to unfiltered URLs used with urllib.request.urlopen, affecting platforms prior to autogpt-platform-beta-v0.6.34. The issue, rated CVSS v4.0 base 9.3 (CRITICAL; NETWORK vector, no user interaction), is patched in v0.6.34. Connected sources (Red...

CVE-2025-62615 AutoGPT has SSRF vulnerability in ReadRSSFeedBlock

AutoGPT is a platform that allows users to create, deploy, and manage continuous artificial intelligence agents that automate complex workflows. Prior to autogpt-platform-beta-v0.6.34, in RSSFeedBlock, the third-party library urllib.request.urlopen is used directly to access the URL, but the inpu...

AutoGPT 代码问题漏洞

AutoGPT is an open-source tool developed by AutoGPT. It aims to make AI accessible and usable for everyone. Previous versions of AutoGPT, including autogpt-platform-beta-v0.6.34, had code vulnerabilities. These vulnerabilities stemmed from the use of the third-party library...

Researchers Uncover Chrome Extensions Abusing Affiliate Links and Stealing ChatGPT Access

Cybersecurity researchers have discovered malicious Google Chrome extensions that come with capabilities to hijack affiliate links, steal data, and collect OpenAI ChatGPT authentication tokens. One of the extensions in question is Amazon Ads Blocker ID: pnpchphmplpdimbllknjoiopmfphellj, which...

CVE-2026-24780

AutoGPT is a platform that allows users to create, deploy, and manage continuous artificial intelligence agents that automate complex workflows. Prior to autogpt-platform-beta-v0.6.44, AutoGPT Platform's block execution endpoints both main web API and external API allow executing blocks by UUID...

CVE-2026-24780

CVE-2026-24780 affects AutoGPT Platform prior to v0.6.44. An authenticated user can trigger remote code execution by calling the execute endpoint for blocks (both main web API and external API) without honoring the disabled flag for BlockInstallationBlock, which writes arbitrary Python code to th...

CVE-2026-0762

GPT Academic streamdaas Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Interaction with a malicious DAAS server is required to exploit this vulnerability but...

CVE-2026-0764

GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific flaw exists within...

CVE-2026-0764

GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific flaw exists within...

CVE-2026-0763

GPT Academic runinsubprocesswrapperfunc Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific...

CVE-2026-0764

GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific flaw exists within...

CVE-2026-0764 GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability

GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific flaw exists within...

CVE-2026-0764 GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability

GPT Academic upload Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific flaw exists within...

CVE-2026-0763 GPT Academic run_in_subprocess_wrapper_func Deserialization of Untrusted Data Remote Code Execution Vulnerability

GPT Academic runinsubprocesswrapperfunc Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific...

CVE-2026-0763 GPT Academic run_in_subprocess_wrapper_func Deserialization of Untrusted Data Remote Code Execution Vulnerability

GPT Academic runinsubprocesswrapperfunc Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific...

CVE-2026-0763

GPT Academic is affected by CVE-2026-0763 via the run_in_subprocess_wrapper_func deserialization flaw, enabling remote code execution with root privileges. The issue stems from insufficient validation of user-supplied data, allowing deserialization of untrusted data, and does not require authenti...

CVE-2026-0763 GPT Academic run_in_subprocess_wrapper_func Deserialization of Untrusted Data Remote Code Execution Vulnerability

GPT Academic runinsubprocesswrapperfunc Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Authentication is not required to exploit this vulnerability. The specific...

CVE-2026-0762

GPT Academic streamdaas Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of GPT Academic. Interaction with a malicious DAAS server is required to exploit this vulnerability but...

CVE-2026-0762

CVE-2026-0762 affects GPT Academic's stream_daas deserialization of untrusted data, enabling remote code execution. The flaw arises from improper validation of user-supplied data, allowing an attacker who can contact a malicious DAAS server to trigger deserialization and execute code with root pr...