AWS Certified AI Practitioner (AIF-C01) — Question 250

A company is using a large language model (LLM) to create a generative AI assistant. The company must choose an AI technique to ensure that the AI assistant generates the most factually correct responses. The company selects the Retrieval Augmented Generation (RAG) technique.

Which limitation of LLMs is the company trying to reduce?

Answer options

• A. Hallucinations
• B. Security
• C. Nondeterminism
• D. Interpretability

Correct answer: A

Explanation

The correct answer is A, as hallucinations refer to instances where the LLM generates incorrect or fabricated information. By using the Retrieval Augmented Generation (RAG) technique, the company aims to reduce these inaccuracies. The other options, such as security, nondeterminism, and interpretability, are not specifically addressed by the RAG approach in relation to factual correctness.