I recently completed the CompTIA AI Essentials course; mainly to understand how close my current use of Generative AI aligns with industry best practices.
New learning I had not previously fully considered:
- Automating daily work tasks: I had initially not been open to the idea of automating sensitive tasks like writing email drafts; however, the course provided a fresh take on how I can improve efficiencies in writing scripts, reports or learning materials for peers.
- More ways to specify outputs: Adding a persona to your prompt can always help narrow down the scope for your answer, tailored to the audience for your output.
Best practices I was already following (after much trial and error over the past year):
- Simple web searches still matter: The simpler the query, it's likely that a simple web search is your best bet; avoiding unnecessary time on generating outputs, writing prompts and fact checking the response.
- Navigating GenAI hallucinations and fact checking: Either through testing with information I already knew, or through fact checking with external sources, it was very evident from the start (and still is) that AI was never a magic spell; it requires careful fact-checking, follow up and clarification to get closer to an expected answer, which we would still need human review and rewrite as the perfect answer.
- Efficient prompting: It is always best to treat GenAI tools conversationally; never expect the correct answer outright and challenge everything in the output, either with the AI itself or compared with external sources.
- Compliance and data privacy: I have always been careful to exclude or replace any sensitive information in my prompt. You can never predict data leaks.
This is a first step on a journey to understand generative AI better. In my current role in cybersecurity, where efficiency and time sensitivity are paramount, GenAI has great potential in improving our scope and efforts on analysis, verification and reaction speed.
As AI continues to evolve, I think the challenge (and opportunity) lies in balancing innovation with responsibility; leveraging AI's strengths while ensuring security and integrity remain at the core.
I'm curious to hear from others: how are you integrating Generative AI into your industry or workflow responsibly?
Originally published on LinkedIn.