220-1102 Question 816
Single answer4.10 Explain basic concepts related to artificial intelligence (AI).A small company wants to deploy an AI assistant to help technicians summarize support tickets and draft email responses to users. During planning, the IT manager says, "If the tool sounds confident, we can trust its answers without review." A senior technician disagrees and recommends a policy that reflects how AI should be used in a support environment. Which of the following is the BEST recommendation?
- A
Allow the AI assistant to send all responses directly to users because AI-generated text is based on training data and is therefore reliable
- B
Use the AI assistant only for repetitive tasks such as summarizing notes and drafting responses, but require human review before sending information to users
- C
Avoid using the AI assistant unless the company can confirm the model stores no training data anywhere in memory or logs
- D
Treat the AI assistant the same as a deterministic script because identical prompts will always produce identical outputs
Show answer and explanation
Correct answer: B
Explanation
The best answer is to use AI as a support tool, not as an unsupervised authority. A+ Core 2 expects candidates to understand basic AI concepts such as the difference between traditional scripted automation and AI-generated output, as well as the need for validation, privacy awareness, and appropriate use cases. In a help desk scenario, AI is well suited for tasks like summarizing ticket histories, suggesting draft responses, or helping organize information. However, technicians should not assume AI output is correct simply because it sounds fluent or confident. Industry best practices from major vendors and security guidance consistently emphasize human oversight, validation of generated content, and careful handling of sensitive data when using AI tools.
- A. Incorrect.
This is incorrect because AI-generated output can be convincing while still being inaccurate, incomplete, or inappropriate for the situation. A common misconception is that a confident tone indicates correctness. In real support environments, sending unreviewed AI responses could lead to misinformation, privacy issues, or poor customer communication.
- B. Correct.
This is correct because it applies AI appropriately as an assistive tool for productivity while recognizing its limitations. In an A+ Core 2 context, technicians should understand that AI can help automate routine work, but outputs should be validated by a human before they are shared externally or used for decision-making. This recommendation reflects practical best practices for responsible AI use.
- C. Incorrect.
This is incorrect because it sets an unrealistic requirement and does not represent the best recommendation in this scenario. Organizations should evaluate data handling, retention, and privacy controls, but requiring proof that no data is stored anywhere in memory or logs is not a practical standard for deployment planning. The more important concept here is to use AI with proper review and data governance.
- D. Incorrect.
This is incorrect because AI systems are not the same as deterministic scripts. A deterministic script follows fixed logic and should produce the same result for the same inputs, while many AI systems are probabilistic and can vary in wording or output. Someone might choose this option if they confuse traditional automation with generative AI.