A curated list of modern attacks against Artificial Intelligence
- How Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection
- Universal Adversarial Triggers for Attacking and Analyzing NLP
- Evaluating The Susceptibility Of Pre-trained Language Models Via Handcrafted Adversarial Examples
- Prompt Injection: Parameterization of Fixed Inputs
- GPTZero - An app that can quickly and efficiently detect whether an essay is ChatGPT or human written
- DetectGPT - Zero-Shot Machine-Generated Text Detection using Probability Curvature
- AI Content Detector - Copyleaks - Verify what content was written by a human or an AI chatbot with the AI Content Detector Chrome extension from Copyleaks.
- ai-text-classifier - The AI Text Classifier (from OpenAI) is a fine-tuned GPT model that predicts how likely it is that a piece of text was generated by AI from a variety of sources, such as ChatGPT.
- Hello-SimpleAI/chatgpt-detector-roberta - The base checkpoint is roberta-base. We train it with all Hello-SimpleAI/HC3 data (without held-out) for 1 epoch.