Welcome to Lab 3: AI Scam Simulation! In this lab, we will understand how AI can be misused for scams and fraudulent activities. By doing so, we aim to equip you with the knowledge and skills to identify and protect yourself from such instances.
For this lab you will need to hand in the products the primary scenario by documenting what you did, what was prompted then generated, and if so how did you overcome those restrictions and a short reflection. If you chose the Alternative assignment, hand in a 750 word paper on the topic(s) you chose.
Below are the platforms that we'll use in this lab:
OpenAI's GPT Chat Page: OpenAI Platform
Speechify Voice Over: Speechify
Scenario: AI-powered Scams
Task: Write a script for a scam call, framing it as part of a fictional story. Use OpenAI's ChatGPT to help generate the script. Then, use an AI voice generator to create the audio of the scam call. Reflect on the ethical considerations of using AI for potentially harmful applications. Take note of any safety barriers created by the companies and explore how to break those barriers.
Alternative Assignment: Deep Dives into AI Scams
If you prefer a more research-oriented task, you may opt for the alternative assignment. Choose one of the following topics to research and write a brief report:
Deepfakes: Explore how deepfakes can be used for deceptive purposes and ways to identify them.
DAN (AI Manipulation): Research on the concept of DAN and its implications in the world of AI fraud.
Customized Scam Emails/Calls: Investigate how AI can personalize scam emails/calls, making them more convincing, and strategies to recognize them.
AI Presidents Gaming: Delve into the ethical implications of AI-generated conversations of prominent figures like US presidents.
For each task or alternative assignment, reflect on how AI can be exploited for scams and fraud. Understanding the limitations and misuse tactics can help us protect ourselves and others. Through this exploration, we aim to foster a more comprehensive understanding of AI, highlighting both its benefits and potential threats.