AI
-
Artificial Intelligence (AI) Big Model Security Risks and Defense In-Depth Report
This report is based on the five core attack surfaces consisting of AI AI critical links from AI Assistants, Agents, Tools, Models, and Storage, with targeted security risks, defense architectures, and solutions.
-
AI zero-hit vulnerability: can steal Microsoft 365 Copilot data
Aim Security has discovered the "EchoLeak" vulnerability, which exploits a design flaw typical of RAG Copilot, allowing an attacker to automatically steal any data in the context of M365 Copilot without relying on specific user behavior. The main attack chain consists of three different vulnerabilities, but Aim Labs has identified other vulnerabilities during its research that may enable exploitation.
-
AIGC Artificial Intelligence Safety Report 2024
Significant progress has been made in the field of AIGC (AI Generated Content). However, technological advances always come with new challenges, and security issues in the AIGC field have come to the fore. The report will deeply analyze the security risks of AIGC and propose solutions.
-
Burpsuite links artificial intelligence (AI) to automate vulnerability analysis
BurpGPT is an artificial intelligence-driven vulnerability detection tool that redefines web application security. Its passive scanning capabilities, customizable alerts, and seamless integration with Burp Suite enable security professionals to accurately and efficiently detect vulnerabilities. By leveraging OpenAI's GPT model, BurpGPT provides comprehensive traffic analysis that goes beyond traditional scanning methods. Embrace the future of vulnerability assessment with BurpGPT and strengthen your network security defenses against evolving threats.
-
Data security: How does generative AI deal with security risks and challenges?
The development and application of AI are having a major impact on the scientific and technological field and may trigger a new productivity revolution. As a powerful technology, AI gives computer systems the ability to generate human language content.
-
Sixteen countries around the world jointly release guidelines for the development of safe artificial intelligence systems
Guidance for any system provider using artificial intelligence (AI), whether those systems are created from scratch or built on tools and services provided by others.
-
AsyncRAT malware analysis practice using Chatgpt
Explore how ChatGPT can assist in analyzing malware, specifically the Remote Access Trojan (RAT) AsyncRAT, and delve into ChatGPT's ability to discuss how it can assist in identifying threat indicators by analyzing network traffic and revealing command and control (C2) infrastructure.