AI | writeups.xyz

Title	Vulnerabilities	Programs	Authors
Unmasking Harmful Content in a Medical Chatbot: A Red Team Perspective	AI LLM Jailbreak Chatbot	Undisclosed	William Wallace (@Phyr3wall)
Microsoft Copilot: From Prompt Injection to Exfiltration of Personal Information	AI LLM Prompt Injection	GitHub (Copilot)	Johann Rehberger (Wunderwuzzi23)
Google AI Studio: LLM-Powered Data Exfiltration Hits Again! Quickly Fixed.	AI LLM Prompt Injection	Google (AI Studio)	Johann Rehberger (Wunderwuzzi23)
From MLOps to MLOops: Exposing the Attack Surface of Machine Learning Platforms	AI RCE XSS Missing Authentication Container Escape Malicious AI Model Malicious Datasets	Jupyter Hugging Face MLflow KServe Seldon	Ori Hollander Shachar Menashe Natan Nehorai Uriya Yavnieli
AI Under Siege: Discovering and Exploiting Vulnerabilities	AI Prompt Injection CSRF Markdown Injection OAuth	Undisclosed	Mosaad Sallam (@H0tak88r) Mohamed Walid (@L0daW)
Unveiling Remote Code Execution in AI chatbot workflows 💵	AI Chatbot RCE	Undisclosed	Anurag__Verma
Jailbreak of Meta AI (Llama -3.1) revealing configuration details	AI LLM Prompt Injection LLM Jailbreak	Meta / Facebook (Llama)	Kiran Maraju
Zeroday on Github Copilot	AI LLM Prompt Injection	GitHub (Copilot)	Marlon Fabiano (@Astrounder)
SAPwned: SAP AI vulnerabilities expose customers’ cloud environments and private AI artifacts	AI Cloud Kubernetes Privilege Escalation Missing Authentication	SAP	Hillai Ben-Sasson (@Hillai) Shir Tamari (@Shirtamari) Nir Ohfeld (@Nirohfeld) Sagi Tzadik (@Sagitz_) Ronen Shustin (@Ronenshh)
Shelltorch Explained: Multiple Vulnerabilities in Pytorch Model Server (Torchserve) (CVSS 9.9, CVSS 9.8) Walkthrough	AI LLM RCE SSRF Insecure Deserialization Zip Slip Attack Path Traversal	PyTorch AWS Google Meta TorchServe SnakeYAML	Gal Elbaz Uri Katz Guy Kaplan Avi Lumelsky
Sorry, ChatGPT Is Under Maintenance: Persistent Denial of Service through Prompt Injection and Memory Attacks	AI LLM Prompt Injection DoS	OpenAI (ChatGPT)	Johann Rehberger (Wunderwuzzi23)
When Prompts Go Rogue: Analyzing a Prompt Injection Code Execution in Vanna.AI	AI LLM RCE Prompt Injection	Vanna.ai	Natan Nehorai
Probllama: Ollama Remote Code Execution Vulnerability (CVE-2024-37032) – Overview and Mitigations	AI RCE Path Traversal Arbitrary File Write	Ollama	Sagi Tzadik (@Sagitz_)
Preauth RCE on NVIDIA Triton Server	AI RCE Arbitrary File Write Memory Corruption	Nvidia	Zhiniang Peng (@Edwardzpeng)
GitHub Copilot Chat: From Prompt Injection to Data Exfiltration	LLM AI Prompt Injection Data Leak	GitHub (Copilot Chat)	Johann Rehberger (Wunderwuzzi23)
Dumping a Database with an AI Chatbot	AI LLM Chatbot	Undisclosed	Kuldeep Pandya (@Kuldeepdotexe)
The risk in malicious AI models: Wiz Research discovers critical vulnerability in AI-as-a-Service provider, Replicate	AI Malicious AI Model RCE Lateral Movement TCP Injection	Replicate	Shir Tamari (@Shirtamari) Sagi Tzadik (@Sagitz_)
Google AI Studio Data Exfiltration via Prompt Injection - Possible Regression and Fix	LLM AI Prompt Injection Data Leak	Google (AI Studio)	Johann Rehberger (Wunderwuzzi23)
Wiz Research finds architecture risks that may compromise AI-as-a-Service providers and consequently risk customer data; works with Hugging Face on mitigations	AI Malicious AI Model Cloud CI/CD RCE Insecure Deserialization Privilege Escalation Supply Chain Attack Cross-Tenant Vulnerability	Hugging Face	Shir Tamari (@Shirtamari) Sagi Tzadik (@Sagitz_)
From ChatBot To SpyBot: ChatGPT Post Exploitation	AI LLM XSS Data Exfiltration Post-Exploitation	OpenAI (ChatGPT)	Ron Masas (@RonMasas)
Security Flaws within ChatGPT Ecosystem Allowed Access to Accounts On Third-Party Websites and Sensitive Data	AI LLM Account Takeover OAuth	OpenAI (ChatGPT) PluginLab.AI KesemAI	Aviad Carmel (@AviadCarmel)
New Google Gemini Vulnerability Enabling Profound Misuse	AI LLM LLM Jailbreak Indirect Prompt Injection Prompt Leaking	Google (Gemini)	Kenneth Yeung
We Hacked Google A.I. for $50,000	LLM AI GraphQL DoS IDOR CSP Bypass	Google	Roni Carta (@0xLupin)
XSS Marks the Spot: Digging Up Vulnerabilities in ChatGPT	AI LLM XSS CSP Bypass Samesite Cookie Bypass Mass Assignment	OpenAI (ChatGPT)	Ron Masas (@RonMasas)
ChatGPT Account Takeover - Wildcard Web Cache Deception	AI LLM Web Cache Deception Account Takeover Path Traversal URL Parsing Issue	OpenAI (ChatGPT)	Harel (@H4r3l)

Page 1 of 2