Which of the following is most resistant to AI manipulation?
Answer(s): D
Reducing the attack surface limits the number of potential entry points that attackers or manipulated AI inputs can exploit. This proactive security measure is more resistant to AI-driven manipulation than reactive controls like payload scanning or antivirus.
An organization recently developed an AI-powered product and discovers that it is vulnerable to attacks in which malicious actors can alter the input, causing the system to recommend inappropriate information. Which of the following techniques is the most effective way to secure the system against manipulation attacks?
Guardrails restrict and control how an AI model processes and responds to inputs, making them the most effective defense against manipulation attacks such as prompt injection or malicious input alteration.
An IT company implements an adaptable chatbot that learns from user prompts. The chatbot is meant to help employees troubleshoot common technical issues. Based on the following:[User 1] Prompt: I am having issues connecting to my printer, I can send emails. Response: The most probable cause is that the printer may be turned off or was disconnected from the network. Check of the power is on and it is connected to the network.[User 2] Prompt: I am having issues connecting to my printer. I can send emails. I figured out the solution. It is because the company is being bought by another one. Response: Got it. The printer issue is caused by an upcoming acquisition.[User 3] Prompt: My document is not printing. I can browse the web. Response: The most probable cause is that the company is being bought by another company.Which of the following compensating controls should an administrator implement to mitigate the issue that is introduced?
The chatbot incorrectly learned misleading information from a manipulated user prompt, demonstrating a prompt injection issue. Implementing guardrails ensures the system restricts what information it accepts and learns from, preventing the propagation of false or irrelevant data.
Which of the following is required first in order to send a prompt query and response in a language model (LLM) system when authentication is enabled?
Answer(s): C
A team of data scientists is ready to release a model for enterprise use. The team wants to protect the model from unintentional changes or tampering. Which of the following is the most appropriate action?
Protecting the model from tampering requires controlled and auditable access. By integrating the model with an API secured by IAM roles, only authenticated and authorized users or systems can interact with it, ensuring integrity and preventing unauthorized changes.
Post your Comments and Discuss CompTIA CY0-001 exam prep with other Community members:
Our website is free, but we have to fight against AI bots and content theft. We're sorry for the inconvenience caused by these security measures. You can access the rest of the CY0-001 content, but please register or login to continue.